ENERDAY, Dresden 11 April 2008

Gas Transportation, Geopolitics and Future Market Structure

Franz Wirl, Yuri Yegorov

University of Vienna, Department of Industry, Energy and Environment, Brunnerstrasse, 72, A-1210,

Vienna Austria; E-mails: franz.wirl@univie.ac.at; yegorov60@gmail.com

March 12th, 2008

Abstract. Energy issues with respect to diversification of European gas imports and Russian gas exports play a substantial role in political discussions of today. We see the core problem in the role of corresponding infrastructure that stresses the spatial component in economic models. Gas market differs from other markets due to very high share of transport and infrastructure costs. The first goal of this paper is to analyse long run gas game that may emerge in the future. Formally, we consider only large participants. In this period 1 there are 3 main gas importers (EU, USA and Core Asia). As for gas exporters, we have broad oligopoly in period 1 (middle term, until around 2030) and narrow oligopoly (Russia and Middle East) afterwards (period 2). All the players have a vector of strategic choices over intensity of exploitation and shares of investment in different infrastructure. In period 1 the producers have to choose their commitment to location-specific pipelines and to shares of LNG in future gas export. Russia can choose between pipelines to EU, to Asia and LNG (to access American market). Therefore, EU might have an interest to lock-in Russian supply via corresponding pipeline investment. Then the long run future gas trade will include 3 clusters: Russia - EU, Iran - Asia, Qatar – USA (monopolistic-monopsonistic relationship with possibly different prices). Symmetric, oligopolistic equilibrium can also exist. We also discuss spatial pricing of gas that is justified by high infrastructure cost and spatial difference in demand. Via changes in internal accounting and transparency of Gazprom as well as introduction of new gas taxing scheme (royalty, export tax) in Russia it would be possible to reduce or even eliminate transit games.

Keywords: natural gas, reserves, trade, European Union, scenarios.

JEL Classification: F14, L95, Q40, Q48.

1. Introduction

Energy issues in general and in particular with respect to Russian gas exports play today a

crucial role in political discussions covering a substantial amount of space in the press. While

being an economic issue, gas trade policy became increasingly politicised in 2006, following

temporal interruption of gas supply to EU in January 2006 that was caused by failure in

reaching price agreement with transit country Ukraine.

The goal of present study is to draw some light on the weakness of EU energy diversification

policy. It is done by the analysis of trends in regional gas production and consumption taking

1

into account the distribution of world reserves. We arrive to the conclusion that after some

time moment (close to 2030 if new discoveries would be low) only few gas exporters will

remain, and the market structure will represent oligopolistic-oligopsonistic game. This is the

main difference of this paper from our report [13], where the main focus was on Russian-EU

relations with respect to gas.

In March 2007, EU Council has adopted new energy policy for Europe. The document

focuses on the necessity to ensure security in energy supply. If energy trends remain, Europe

will be much more dependent on energy imports: by the year 2030, 84% of gas and 91% of oil

will be imported. It has an ambitious objective that Europe to lead new industrial revolution

that will reduce greenhouse emissions by 30%, improve energy efficiency by 20%, raise the

share of renewable energy to 20% by the year 2020 [9]. OPEC World Oil Outlook [10] is

more pessimistic about the possibility to slow down consumption of oil and gas so quickly.

Europe’s declared strategy for energy (and gas) diversification is understandable. However, it

should take into account the evolving temporal pattern of gas suppliers. Here we have to

distinguish between middle and long term horizon. In the middle term (by the year 2020) it is

important to consider existing infrastructure projects. This approach has been taken by

Jonathan Stern (2006). The decline in production and rising import dependence of EU is

obvious and more detailed analysis will follow.

Sufficient gas reserves exist in a range of regions: Russia, Middle East, Northern Africa,

Caspian region. While in the middle run planned and existing infrastructure determines the

future, in the long run it is the ratio of production to reserves that will determine gas supply.

Our paper will take this factor as a core argument. Due to resource availability and spatial

aspects, this EU diversification must be restricted to African and Arab countries and Iran and

will mostly be in the form of LNG. Since the time span of these sources1 is rather short (say

20-30 years), Western Europe will, whether it likes or not, have to turn on Russia and Middle

East in around 30 years due to their substantial gas resources.

On the other side, Russia may want to increase its natural gas exports in the near future. In

this respect, expanding exports to Western Europe seems the natural option. However, facing

obstacles for marketing in Western Europe, it may turn to other options: the Chinese, Pacific

and the American market. The major thesis of this paper is to highlight the trade off between

1 African and European gas resources are restricted in their availability in the long run given the targeted rate of exploitation while the development of the huge supplies from Iran faces political obstacles.

2

Europe (i.e. the EU) short term concern about energy security via diversification versus long

term supply availability of natural gas from Russia, because Europe’s diversification strategy

may induce Russia to seek other markets.

Major features of this study are: First, the long term perspective for gas market, in particular

taking the resource limits of gas suppliers into account. Second, peculiarities of the gas

market, with the oligopolistic supply and the oligopsonistic demand (first characterized in

Golombek et al. (1987)), when the spatial aspects and the sunk investment into infrastructure

(pipelines) are explicitly taken into account.

Section 2 reviews the gas market concluding that alternative suppliers of natural gas for

Europe have a rather limited time window due to their shrinking and limited resource basis.

Section 3 sketches Europe’s strive for energy security via diversification and lists the options

in the light of the analysis in Section 2. In Section 4 we consider possible evolution of future

gas market and formulate possible market structures and corresponding games. Section 5

sketches scenarios, or more precisely, different options how Russia will and could expand its

natural gas industry. Section 6 seeks to obtain what are the implications of our analysis

accounting for economic as well as political objectives.

2. The Natural Gas market

The following discussion uses data from British Petroleum (Davies, 2006). Since regional

distribution of additional resources is unknown, the following discussion is based on the

proven reserves of natural gas, which will affect our conclusions only with respect to

stretching the time scale, if the distribution of additional discoveries follows those of the

existing reserves2.

2.1. Demand, supply and reserves

The world consumption and production of gas is rapidly growing. If the trends of gas

consumption would remain, the demand for imported gas will grow very fast in the USA and

Pacific Asia (mostly driven by growing China). Under these conditions, even modest growth

2 Indeed, investigators differ even for what are called ‘proven’ reserves. For example, the world gas reserves in 2005 are estimated at two levels: 6043 trln.cub.ft (Oil and Gas Journal), or 6997 trln. cub.ft (World Oil), http://www.eia.doe.gov/emeu/aer/txt/ptb1104.html

3

of European gas demand would put Europe into fierce competition for gas supplies with other

main gas consumers. The possibilities of diversification of gas supply look favourable for the

next 2 decades, but will rapidly erode with the extinction of gas in the countries that have

moderate reserves (like Norway and Algeria). After that Russian role as gas supplier will

become much more important, and the main Russian competitors in supply would be in the

Middle East. Such diversification is much smaller in comparison to what is seen feasible

today.

Despite the development of alternative energy sources, gas and oil would remain primary

energy sources for the next 20-30 years. ExxonMobil3 estimates the annual growth of world

demand for energy during 2000-2020 at 1.7%, with the demand for gas growing at the

average annual rate of 2.4%. The main driving force for its growth is power generation, where

the sectoral demand for gas will grow at 3.6%.

Table 1. The main world gas reserves holders, their production and consumption. Source: BP data [1] and calculations by authors.

N Country Reserve % R/P

yearsProduction

mtoe YUSCons. 2005

Export mtoe Export%

1 Russia 47,82 26,6 80 538,2 75,52 364,6 173,6 32,3 2 Iran 26,74 14,9 >100 78,3 42,23 79,6 -1,3 -1,7 3 Qatar 25,78 14,3 >100 39,2 40,72 14,3 24,9 63,5 4 Saudi Arabia 6,9 3,8 99 62,6 10,87 62,6 0 0,0 5 OAE 6,04 3,4 >100 41,9 9,54 36,4 5,5 13,1 6 USA 5,45 3 10,4 473 8,63 570,1 -97,1 -20,5 7 Nigeria 5,23 2,9 >100 19,6 8,26 n/a n/a n/a 8 Algeria 4,58 2,5 52,2 79 7,23 21,7 57,3 72,5 9 Venezuela 4,32 2,4 >100 26,1 6,82 26,1 0 0,0

10 Iraq 3,17 1,8 >100 n/a 5,01 n/a n/a n/a 11 Kazakhstan 3 1,7 >100 21,1 4,74 16 5,1 24,2 12 Indonesia 2,76 1,5 36,3 68,4 4,36 35,5 32,9 48,1 13 Australia 2,52 1,4 67,9 33,4 3,98 23,1 10,3 30,8 14 Malaysia 2,48 1,4 41,4 54 3,92 31,4 22,6 41,9 15 Norway 2,41 1,3 28,3 76,5 3,80 4 72,5 94,8 16 China 2,35 1,3 47 45 3,71 42,3 2,7 6,0

The Table 1 shows the main 16 gas reserve holders in the world, together with their R/P and

production levels in 2005 (in Mtoe). The exploitation of gas reserves is quite heterogeneous

across the countries and regions with P/R ratios ranging from 10 years in North America to

240 years in the Middle East.

3 Luxbacher Robbie, director Exxon Mobile Gas and Power Marketing “The Growing Importance of Natural Gas Industry”. – Presentation at the Oil and Money Conference in London, October 26, 2004.

4

However, the rate of exploitation is likely to change over time. If the forecast of ExxonMobil

[2] proved to be correct, the intensity of exploitation in Africa would grow dramatically in

2010-2020, exceeding the present export of Russia and still above the level of Middle East. At

such speed African reserves are sufficient only for about 20 years. The present relation

between output and consumption of gas for leading gas producers is shown in Fig.1.

Fig.1. Gas production and consumption 2005 (mtoe)

Source: BP

0

100

200

300

400

500

600

Russia USA

Canad

a UK

Algeria Ira

n

Norway

Indon

ezia

Saudi

Arabia

Malays

ia

Uzbek

istan

China

OAEQata

r

Austra

lia

Venez

uela

Kazak

hstan

Nigeria

Ukraine

German

y Ita

lyJa

pan

Countries - leading producers

Volu

me,

mto

e

Prod 2005Cons 2005

2.2. Changing pattern of world gas reserves

We want to reveal the changing structure of the world pattern of gas reserves. According to

the reference scenario from World Energy Outlook 2006, the annual growth rate of gas

production will differ across regions. Between 2004 and 2030, it is expected that Middle East

and Africa will have the highest growth (4.5%), followed by Latin America (4.1%). Although

transition economies (where Russia plays the major role) are expected to have lower growth

of gas production (1.3% per year), in 2030 their produced gas volume will still remain the

largest among the regions (above 1100 bcm/year), followed by Middle East (see Fig.4.3 from

WEO2006[12]). The production of gas in OECD Europe will decline at 0.5% per year, while

gas production in North America will grow at 0.4% (with output in 2030 being comparable

with Middle East, at 800-900 bcm). The growth in gas demand will also be different across

world regions. Gas demand from developing countries is expected to grow at 3.7% per year

(with China showing the highest growth at 5.1%), while EU demand will grow only at 1.4%,

slightly above OECD demand (1.2%) and demand from transition economies (1.3%).

5

One of our goals is to check the long-run sustainability of gas diversification policy by EU. It

is necessary to make several additional assumptions. First of all, it is assumed that countries

from region have the same growth rate in production and consumption as their region.

Second, it is assumed than no new gas will be discovered4. Figure 1 shows the dynamics of

gas reserves for smaller reserve holders (excluding Russia and Middle East). We see that if

production growth will develop as predicted by IEA, Algeria and Norway (main present

competitors of Russia for gas supply in EU) will have practically no reserves. The same will

happen to main Asian producers (Indonesia, Malaysia) and Australia, who supply Japan and

Korea with gas at present. At the same time China will stay without own gas and with its

fastly growing demand will become one of the main competitors among gas importers at the

world market. Only Nigeria, Venezuela and Kazakhstan will have substantial reserves. By

year 2030, only Kazakhstan can be important EU supplier among this group, since USA will

submit high bid for Nigeria given than it may not get easily gas from Venezuela given

political reasons and growing demand for it from Latin America.

Fig.2. Dynamics of gas reserves for smaller reserve holders (excl.Russia and Middle East). Source: BP data, IEA scenario. Calculation by authors.

-1

0

1

2

3

4

5

6

Nigeria

Algeria

Venez

uela Iraq

Kazak

hstan

Indonezia

Australi

a

Malays

ia

Norway

China

Res

erve

s, tr

ln.c

um R2005R2010R2015R2020R2025R2030

4 In a complementary scenario we can assume that there will be overall discoveries equal to the present reserves

(like it is assumed in the mean scenario of American geological agency) and that they will be added

proportionally to existing reserves.

6

2.3. Reserve-Production Ratio and Discoveries

gy 2006 [1], reserve-production ratio (R/P)

for gas in 2005 was 65.1 years, down from its maximum in 2000 (70 years). New discoveries

tates that the areas

that contain the greatest volumes of undiscovered conventional gas include the West Siberia

.3. Spatial Aspects

ery important for natural gas in contrast to many other commodities

including crude oil since costs and transportation modes depend significantly on distance and

According to BP Statistical Review of World Ener

are also important, since they may change reserves, and do it heterogeneously across regions.

The US Geological Survey (USGS) World Petroleum Assessment 2000 provides estimates of

oil and gas that has the potential to be added to reserves. While these estimates are

probabilistic5, we still can use these means for predicting future reserves (resources). For the

USA, the present gas reserves of the USA are only 3% of the world reserves, while their share

in undiscovered reserves is close to 10%. This means that the situation with gas in the USA is

not so pessimistic as it would be if we just project the impact of their current consumption

trend on reserves. For the world, the mean value of undiscovered gas reserves is

approximately equal to proven reserves, and this doubles the future reserve-production ratio.

This means that there is enough gas in the world not for 65 years, but maybe for 130 years (in

the pessimistic scenario for 70-100 years, and in optimistic for 150-200).

The second question is how this will change across the regions. USGS s

basin, Barents and Kara Sea shelves (this is Russian economic zone) and some regions in

Middle East and offshore Norwegian Sea. Hence, all arguments of our paper change only

quantatively (and not qualitatively) if we take into account changes in reserves due to

undiscovered gas.

2

Spatial aspects are v

geography (land or sea). A crucial decision is to deliver natural gas via pipelines, which

requires enormous sunk and location specific investments, or as liquefied natural gas (LNG)

that requires also huge sunk costs but more flexibility in exporting and importing. A rule of

thumb is that pipeline transport is more efficient below 4000 km if over land, and up to 2000

5 For example, the almost secure (at 95% probability) level of reserve growth for gas (2299 TCF) is about one-half of the mean (4333 TCF at 50% probability).

7

km under sea6. Hence the choice of transportation is consequential providing a partial lock-in

for exporters and importers in particular in the case of pipelines.

Therefore, one must look at the spatial distribution of gas reserves and their location relative

European Union has three main sources of gas at present: from Russia (by transcontinental

2.4. Outlook for the development of regional patterns of trade

The broad picture of future natural gas demand in Section 2.1 is here complemented by

USA - the World. The USA is the largest world gas consumer and still the second largest

to the consumers. Gas industry is a spatial network industry beyond the unequal distribution

of natural gas across the globe. Spatial topology of pipelines and other geographical factors

play an important role here. Transportation cost of gas represents an important price

component, and thus there exist a set of regional prices instead of unique world price.

pipelines with length about 4000 km), from Northern Sea (developed by Norway, Netherlands

and UK) – by sea pipelines, from Northern Africa – delivery by LNG to Mediterranean ports.

regional patterns and a discussion of trade flows supporting these demand projections. A

crucial distinction is in the mode of transportation, either via pipelines or as liquefied natural

gas, short LNG.

producer (see Table 1) although its reserves are only modest (3%). Therefore any change in

the US demand for imports will have a substantial influence on world imports. Given the US

demand for gas and its P/R ratio of 10 years, it will have to increase its future import

substantially. Where will these imports come from? The American continent is less rich in gas

reserves than Eurasia. According to BP Statistics 2006, North America has only 4.1% of gas

reserves, while South America has 3.9%. As a consequence the expansion of imports via

pipelines is limited7, the USA has to rely on substantial increase of LNG imports in future.

The major current US gas supplier, Trinidad and Tobago, has small reserves (0.3% of the

world stock) and thus they cannot meet the US import demand in the long run. The next close

potential supplier is Venezuela, with 2.4% of the world stock, is not likely to become main

6 Abundant World Natural Gas Reserves and Natural Gas Potential (http://www.eia.doe.gov/oiaf/analyisispaper/global/natgas.html). 7 At current, Canada supplies gas to the USA by pipelines, but its own reserves are smaller than those of the USA (0.9% of the world stock with a P/R ratio of only 9 years).

8

supplier for political reasons; it may prefer to build pipelines to fuel the closer Latin American

markets. The USA may look for gas exports from African countries and Middle East.

Russia - Europe. Northern Europe will be more dependent on Russian gas in future8. The

Russia – Asia. At present, the demand of Japan and S. Korea for gas is satisfied by LNG

3. EU: Supply Security and Future Scenarios

The security of gas supply is one of the most important topics for EU nowadays. Weisser9

main reasons are depletion of its own production. Southern Europe can rely more on the gas

from Northern Africa due to its lower delivery cost. But the main problem is the balance

between growing demand for gas in Europe and the growth of exports of gas from Russia.

Russia may not fulfil all European demand for its gas in future due to several reasons: a) too

slow growth of domestic gas extraction due to growing cost of gas extraction development, b)

reorientation of export to other markets (Asia).

supplies. The role of China as gas consumer is growing: in 2005 it consumed 47 bln.cub.m of

gas, contrary to only 23.8 in 2000. There is a technical possibility of gas pipeline from Eastern

Siberia to Far East, with Japan, Korea and China as potential consumers.

describes an increasing role of gas for EU, makes middle run forecasts about gas trade (by

2030, global gas trade will increase from current 22% to 40%, and LNG will form 50% of this

trade) and concludes that “the present gas supply to the EU depends dangerously on too few

sources”. Indeed, the present gas supply to EU depends on too few exporters (with Russia

playing the dominant role), depends on pipeline infrastructure, “the system without

flexibility”, that involves several types of risks (including political dependence on transit

countries and international terrorism). Both new pipelines and LNG infrastructure are very

expensive equipment, and both require long run contracts. This contrasts the beliefs of many

Europeans about the benefits from gas market liberalization. In contrast to this consumer-

oriented approach, OPEC recognized the necessity of security in both energy supply and

8 This is not only authors’ opinion, but also has been suggested in the report: I. Bozon, W. Campbell, T. Vahlenkamp “Europe and Russia: Charting Energy Alliance” – The McKinsey Quarterly: The Online Journal of McKinsey &Co, January 2007. 9 See Weissler H. “The security of gas supply – a critical issue for Europe?” – Energy Policy, 35 (2007), p. 1-5.

9

demand, the necessity to minimize uncertainties along the supply chains, to reduce investment

risks, and to support long-term market stability10.

Huge trans-Eurasian gas pipelines will still be used in the middle run as a bulk supply of

Russian gas to EU. The problem is that doubling infrastructure for competition reasons leads

to spare capacities that are too expensive11. Even if an investor will emerge to build a

competing network of gas supply to Europe, it will charge much higher prices than present.

If we look at the map of existing gas infrastructure we can see what realistic options EU has

to increase its gas security in the short and middle run. There are LNG receiving facilities,

mainly located in Mediterranean countries and having LNG gas supply from Northern Africa.

Given the small distance and lack of Trans-European gas pipelines to Iberian countries, the

shift to LNG seems the proper economic decision. But one should also look who will export

LNG. If we look at the countries of Northern Africa and the BP Table for gas reserves, we

will see that the main player here is Algeria, who also has LNG infrastructure ready. But its

reserves are only 2.5% of the world, and any substantial increase of European gas imports

from Algeria in the short run empties its gas reserves in the medium term. However, existing

infrastructure is only important in the next 10 years, since new facilities can be built.

EU has also its own gas reserves. There are two gas net exporters in Western Europe: the

Netherlands (with 0.8% of world gas reserves and current R/P at 22 years) and Norway (1.3%

of world reserves, R/P= 28 years). But these reserves are even less than Algerian covering

only 10 years of present EU consumption. This means that reliance of EU in its diversification

of gas imports on Western European (Netherlands and Norway; in fact, they play strategic

role in German diversification of gas suppliers today) and Northern African sources would

deplete all these known reserves in about 25 years.

3.1. Gas after 2030

Let us look after the year 2030. In fact, some gas infrastructure being built today depends on

gas flows also after the year 2030. The scenario described above shows that after 2030 Europe

10 Further significant development in the EU-OPEC energy dialogue. – Joint press release, Vienna, 21 June 2007 11 McKinsey estimate that it would cost for EU an additional $40 to $60 bln (net present value) more to meet its energy needs from sources other than Russia. Russia would forgo $50 to $70 bln.(NPV) if its gas from Western Siberia has to be rerouted to Asia (see [4]).

10

Fig. 3. Dynamics of gas imports by main regions

0

100

200

300

400

500

600

700

2010 2020 2030Year

bln.

cum

North AmericaRest Asia & PacificEur, w/o Russia

becomes dependent on other gas producers. Russia still remains in this club, having almost

30% of world gas reserves. Other producers-survivors are located in the Middle East, where

the joint reserves of Iran and Qatar are comparable with Russian.

The following calculations are based on BP data on gas (2005) and scenarios of regional

trends in gas production and consumption from World Energy Outlook (2006). The

calculations show that the division between exporting and importing regions will be growing.

Fig.3 shows that Europe12 is the main gas importing region, and its dependence on gas

imports will be growing. Four regions (Russia, Middle East, Africa and Latin America) are

gas exporters, but their relative role will change over time (Fig.4). Despite the fact that

Middle East holds the main gas reserves, it will remain on the 3rd position, while Africa can

take over Russia in gas export volume between 2020 and 2030. How sustainable will be this

pattern in the long run? It is important to have a look at gas reserves (Fig.5). Here the

pessimistic assumption about no gas discoveries in employed. We see that in 2030 Africa will

loose more than half of its reserves and thus will not be able to maintain its leading exporting

position for many years. Russian reserves will erode faster than those of Middle East, but only

so gas exporters. That is why EU gas

dependence is more pronounced that can be seen from the Figures below.

12 In BP statistics Europe is viewed as Eurasia without Middle East and South-East Asia. Since Russia is gas-

exporting country, it was decomposed in a separate region in Table 1. Europe here contains not only EU, but

also FSU countries excuding Russia. Some of them (like Kazakhstan) are al

11

Fig. 4. Dynamics of gas export by regions. Source: BP data, IAE scenario. Calculation by authors.

0

50

100

150

200

250

300

350

Russia Africa Mid.East C.&S.America

bln.

cum 2010

20202030

these two regions will remain long term players in this game.

Fig. 5. Dynamics of gas reserves by regions (no new discoveries)

-20

-10

0

10

20

30

40

50

60

70

80

NorthAmerica

C.&S.America Mid.East Africa Rest Asia &Pacific

Russia Eur, w/oRussia

Res2005Res2030

12

4. Future Games

From theoretical perspective, it is possible to consider a game with two periods. The period 1

lasts until 2030 (or some time close to it). In this period there are 3 main gas importers: EU,

USA and Core Asia13, which remain main importers also in the 2nd period. As for gas

exporters, we have broad oligopoly in period 1 and narrow oligopoly in period 2. This

depends very much on present policy chosen by EU. Narrow oligopoly contains 3 main

holders of gas reserves: Russia, Iran, Qatar. Due to closeness of geographic locations, we can

join Iran and Qatar into one player: Middle East. All the players have a vector of strategic

choices. Putting it in a simple formal framework, we can say that players choose intensity of

exploitation and shares of investment in different infrastructure. EU can either push for an

expansion of competition or it can commit itself to deal with narrow oligopoly in period 1,

like it will do in period 2. The more is competition in period 1, the more small players start to

deplete their reserves fast, and the quicker will be the period with narrow oligopoly.

4.1. LNG-Pipeline Split of Future Gas Demand

Both pipelines and LNG infrastructure have high fixed costs. Moreover, a commitment to

pipeline is location-specific, contrary to decision of building LNG plant. That is why in period

1 the producers have to choose their commitment to location-specific pipelines and to shares

of LNG in future gas export. If they make such commitments in period 1, they will be not

interested to reverse them in the period 2, due to high cost of building new infrastructure.

The strategy of USA is high bid for LNG. Asia is heterogeneous. While Japan is fully

dependent on LNG now (like USA in future), China is indifferent between pipeline gas and

LNG. As Chinese demand is likely to grow fast, we can think that future Asian gas demand

will be approximately equal split between LNG and pipeline gas. As for Europe, its present

share of LNG is low (less than 15%), but may increase in future. The problem of EU that is

will not be the hardest bidder for LNG in future, since due to geographical properties it is not

as fatally dependent on them as Japan or USA.

13 We will name by Core Asia (later simply Asia) main gas importers in Asia (Japan and China).

13

Now let us consider producers’ strategy versus LNG and pipelines. Let us denote the shares of

particular infrastructure by Russia into European pipeline, Asian pipeline and LNG by a

vector (RE, RA, RL). Similarly, shares of Middle East investment into regional pipelines and

LNG are (ME, MA, ML). For example, if a vector (ME, MA, ML)=(1/3,1/3,1/3), this means

that Middle East fully diversifies its infrastructure investment, and in the long run will supply

1/3 of its gas via European pipeline, 1/3 via Asian pipeline, and 1/3 will trade as LNG. While

Middle East has only started its infrastructure investment (small investment in LNG by Qatar

and no investment by Iran yet), Russian bulk investment is already into European pipelines.

Only its extra capacity (if it chooses to expand it) will split among 3 alternative investments.

Hence, Russian investment vector will be biased to the first component. If Middle East will

diversify fully, i.e. choose (1/3,1/3,1/3), then the long run shares of LNG may be between 1/4

and 1/3, if Russia chooses pipeline investment and even lower if it chooses LNG. Note that in

this case the total amount of produced LNG may be much less than will be demanded by USA

and Japan. In this environment, they will present the highest bids, and EU is most likely to

stay without any LNG. If Middle East would choose only LNG (which is OK for Qatar, but

maybe not the best option for Iran14), then the long run share of LNG in the market can be

50% or above. In this case, fully LNG-dependent USA and half-LNG dependent Asia will be

satisfied, and some LNG may come to Europe.

4.2. Some Statements

Now we will formulate rather intuitive statements, or Propositions, without formal proofs.

Proposition 1. Period 2 will start at some moment.

Propostion 2. Russia is indifferent between constructing more pipelines to EU or building

LNG plants. Pipelines to EU are slightly preferred to pipelines to Asia.

Proposition 3. If LNG becomes scarce, USA will submit higher bid for them than EU.

14 In this case LNG concentration in Persian Gulf will be too high, and this region will be extremely vulnerable

to terrorist attacks with severe consequences for both gas producers and even more for LNG importers.

14

Proposition 4. If Russia builds LNG plants or Asian pipelines in period 1, it is not optimal for

it to increase gas sales to EU by building new pipelines in period 2.

Proposition 5. EU should invest in pipeline to Russia in period 1, to reduce Russian risk in

energy demand and to reduce Russian incentive to invest more in Asian pipelines and LNG.

By doing this, EU can secure its gas supply in period 2.

4.3. Possible Equilibria

There might be different equilibria (even multiplicity) given the details of model

specification. For simplicity, we assume 3 exporters (Russia, Iran, Qatar) and 3 importers

(EU, USA, Asia) that are identical in economic parameters (production function, export

volume, demand functions) but differ only in geography. That is, 1st consumer (US) can

import only LNG, 2nd can import approximately equal mix of pipeline and LNG gas (Asia),

and 3rd consumer can choose LNG fraction between small number and ½. As for producers,

the 1st (Russia) can choose LNG between 0 and some positive number15, while the 2nd and the

3rd can choose almost any portfolio of gas infrastructure investment16.

Equilibria 1 (monopolistic). Suppose that Qatar fully invest in LNG, Iran in pipeline to Asia,

while Russia in pipeline to EU. Then the long run future gas trade will include 3 clusters:

Russia - EU, Iran - Asia, Qatar - USA. Each of these 3 couples will be monopolistic-

monopsonistic relationship.

Equilibria 2 (oligopolistic). This equilibrium can be symmetric. For example, 3 producers

produce mix (1/3,1/3,1/3), and 3 consumers import this mix. It is quasi-competition, and

despite small number of players, prices can be not much different from fully competitive.

15 Russia already invested a lot in pipeline gas infrastructure to EU. Depending on expansion choice, its shares of

LNG and Asian pipelines can be some positive shares, but also zeroes.

16 Both Iran and Qatar have high R/P ratios for gas and high reserves. This means that in the long run they would

definitely increase gas export substantially.

15

5. Russian Strategies and Scenarios

The real life is always more complex than the model. The game above was only about the

selection of investment strategies. However, there are more dimensions of Russian strategies

that have to be characterised17. The first two scenarios correspond to high discount rate (and

thus fast exploitation of gas resources). The difference is only in the dominant choice of

future investment (pipelines vs. LNG). If Russia would reduce uncertainties, its future

discount rate will drop, and it may choose to keep more gas for future generations. In

particular, scenarios 3 and especially 4 assume a much lower time discount than present.

Scenario 1: Fast development of gas production and pipeline capacity. Russia would

increase its pipeline capacity and this will guarantee future supplies to EU. If time discounting

is high (like it has been until today), it is likely to follow this path. The negative side is that

this leads to games with transit countries. To reduce that, more pipelines will be built to

connect Russia and EU directly (Baltic and Black Sea pipelines). This scenario might be the

best for EU in the long run. At present, EU seems to push Russia away from this scenario.

Additional investment in gas infrastructure is costly and given even moderate risk aversion

can be provided only after guarantees of Europe on long term and stable gas imports from

Russia. But that is exactly opposite to what Europe is offering Russia by signing Energy

charter that guarantees security in supply but not security in demand.

Scenario 2: Fast development of gas production and LNG. Russia might be indifferent

between expansion of pipeline network or investment in LNG facilities. This scenario also

corresponds to high time discounting. Contrary to scenario 1, Russia can choose this scenario

if threatened by lack of access, long term contracts by EU and transit games by some

European countries. Under this scenario, Russia makes less commitments and strategically

decides to have a fraction of gas that can be sold to any buyer in the world. The main

consumers of Russian LNG are then likely to be USA and Japan, while Europe is likely to

have less Russian gas in its future portfolio.

Scenario 3. Slow expansion of production; growth of export due to re-export of Central Asian

gas and domestic consumption decline. Russia already started to use Central Asian gas

(mainly from Turkmenistan) for sales to Ukraine and other European countries. Since

17 More details can be found in (Yegorov, Wirl, 2007)

16

Turkmenistan is a land-locked country, it has few options to sell its gas. As a consequence,

the Russian state effectively controls Central Asian reserves because of its dominance of the

transit routes through the Transneft and Gazprom monopolies (see Monaghan (2006)). If

Russia’s domestic discount rate falls, it can start to realize that the possibility not to expand its

gas production is also viable and consistent with an increase of gas revenues. There are two

sources of it: a) decrease of the share of gas that is used domestically (at present its price is

about 20% of the world price), b) exploitation of gas reserves of neighbouring countries, that

have limited possibility for direct gas trade due to location. However, it is not obvious if this

scenario is sustainable in the long run.

Scenario 4: Keeping present steady state in gas production and export to preserve

reserves until they become more scarce. This scenario can occur when the Russian discount

rate declines substantially. The necessary pre-condition is putting inflation and interest rate to

3-5%, like in EU and the USA. The price of gas will definitely increase after 2030, because of

extinction of a significant part of oil and gas reserves.

6. Pricing of Russian Gas

Spatial differences in gas pricing (especially for Russian gas) also became a question of

political debates. It is often claimed that low internal gas prices in Russia and some

intermediate level for post-communist countries is a political game by Gazprom and that it

can be eliminated through market liberalization. The goal of this section is to draw some light

on price formation process, multiplicity of possible market structures, externalities, role of

internal accounting and social welfare. Since an article section is too little space for covering

this agenda in more detail, only some statements will be made, with more elaboration planned

in further articles.

The first important question is about gas price formation. Besides standard market

considerations (a la neoclassical economics), there are two more issues that play very

important role. The first is related to spatial economy, and the second to economics of

exhaustible natural resources. It was Hotelling, who made contributions to both issues about

80 years ago. Despite the claims of the school of new economic geography about vanishing

role of physical space in modern globalized world, the situation with gas is different. It is

determined by very high share of transport costs in the total cost (gas transportation is about 8

17

times as expensive as oil) of the product and also by emerging natural monopoly along most

of the pipelines.

Economics of exhaustible natural resources clearly sets the question of owner of the deposits

of those resources, which may be different from producer. According to old Soviet

constitution, resources belong to the whole population. However, privatization in 1990ies did

not raise the question of natural rent and its owner. If natural rent for gas will be introduced in

Russia, this can eliminate some of the questions.

Consider a hypothetical example, however with realistic order of prices. Suppose that at some

moment the European price for gas (per 1000 cum) is $300, Russian internal price is $50, cost

of extraction is $10, cost of domestic transportation varies between $30 and $50 depending on

region, and cost of export infrastructure is $100. In the environment on no natural rent,

producer of gas makes before tax profit of 300-100-10 =190 ($ per 1000 cum) on exported

gas and almost zero profit on domestic sales (in the case of flat prices there is cross-subsidy of

remote regions on the expense of central). Then simple market liberalization will force

producer to raise domestic price.

Now let us consider the effect of natural rent. Suppose that it is set by law at the level of $50

(per th.cum). Then the cost of gas at the point of extraction (we abstract here from

heterogeneity of costs across locations of gas deposits) becomes not $10, but $60. The cost of

delivered gas will vary domestically between $90 and $110, and pre-tax cost to export gas

becomes 10+50+100=160. Suppose now that natural rent is uniformly distributed across the

population. Then raising domestic price to zero profit level of $100 and paying out natural

rent of $50 will make little difference for domestic consumers in comparison with baseline

case of no natural rent and price of $50 (except for incentives to save gas). However, producer

may still be interested to export more gas at the expense of domestic consumption. This can

be eliminated by introduction of export tax. Thus, introduction of two laws (natural rent and

gas tax) can replicate the present price pattern but now correspond to economic equilibrium

(of regulated monopoly type). It is clear that price in this equilibrium will depend on location:

$90-110 inside Russia and $300 in Far Abroad. Since the length of pipelines (and hence,

transport cost) to supply gas to former Soviet republics is at intermediate level, the market

price for them should be less than for EU, but above Russian domestic price (however,

depends on potential difference in export tax for those consumers).

18

This example shows that gas price should have persistent difference across locations because

of transport cost. However, in the environment of natural monopoly (on any segment of

strategic pipeline, due to scale economies) it is difficult to support a unique spatial price

pattern by market equilibrium. Here we should consider how different market structures can

support different spatial price patterns.

6.1. About Strategic Role of Internal Accounting in Gazprom

The final gas price for EU consumer has many components: a) rent of using exhaustible

natural resource (the higher is it, the better it is for resource owner; if resource owned by state,

then maximal increase of this rent becomes national interest), R; b) extraction cost, E; c)

direct competitive transit cost (cost to pump, discounted construction cost, maintenance cost),

TC; d) tax on extraction, transit and export, T1; e) transit surplus (rent of intermediary plus

profit of transit companies), TS; f) delivery and distribution cost inside EU, DC; g) profit of

local EU distributors, PEU; h) EU taxes on gas and distribution, TEU.

Let point A corresponds to location of gas producer (Russia) and point B – to location of larg

consumer (EU). The selling price of Gazprom to EU, P(B), includes R, E, TC, T1 and TS. If

state policy is to maximize R+T1, it should create market structure to set TS=0. If R is

maximized first, it is important to have competitive price implementation in extraction and

transit. If there is simultaneous interest to subsidize domestic consumers of gas, R can be

lower to reduce domestic price, but additional tax revenue should come from export tax. At

present, these issues are not clear because they are inside internal Gazprom accounting.

Rather then to separate the ownership of extraction and transmission, Russian state can

introduce monitoring of accounting inside Gazrpom.

It is possible to have different market structures in Russia for gas. Because of natural

monopoly, perfect competition is impossible (and separation of production from

transportation will not eliminate monopoly power at least at transit segments!), and thus

monopoly should be regulated. Introducing royalties and correct export tax can do more for

efficient equilibrium, while liberalization under present laws may have horrible consequences.

Let us focus on spatial price structures under different markets. If it would be possible to have

perfect competition in gas transportation (this will not happen as very large number of small

pipelines is cost-inefficient), then price for transit is exactly equal to its cost. Due to linearity

19

of transport costs in distance, P(x)-P(y) = T |x-y|, where T – unit distance transport cost. In the

case of monopoly, we know from Chamberlin (1933) that it may be optimal for monopolist

who can spatially price differentiate to charge only half of its transport cost for consumer.

Note that both pricing schemes do not allow for spatial arbitrage. However, the main problem

is to know exactly the level of transport cost (parameter T). This can be done only by internal

accounting and in the case of Russia revealed through audit of Gazprom.

If one monopoly owns both production and transportation, the main problem is internal

efficiency. If there would be transparency in internal accounting of Gazrpom, then regulator

can check whether transportation sub-accounting results in zero profit. Here not only variable

costs but also fixed costs of constructing pipeline (divided by the time of expected use) have

to be added. Internal accounting plus establishing of royalty and export tax levels allows for

producing many outcomes, including current prices, by varying these parameters. That is why

this "regulation equilibrium" may not perturb existing status quo, but can give some legal and

market support to it.

6.2. About Gas Transit Games

Non-transparency of internal accounting data does not allow for setting international law

about acceptable range of prices for transit. The recent gas game between Russia and Ukraine

(February-March 2008) has revealed clearly that “transit price” has little to do with transit

cost, but represents rent seeking. That is why Gazprom is so eager to pursue with Nord

Stream and South Stream projects, despite higher level of construction costs comparing to

land pipelines. In particular, Ukraine suggested 5-fold increase of transit price, from $1.70 to

$9.32 for 1000 cum per 100 km [14]. To see that such values have nothing to do with cost, it

is enough to calculate that application of the same rule for transiting Central Asian gas to

Ukraine will make it more expensive than in Western Europe (only transit cost will be close

to $300 for more than 3000-km long pipeline). Russian replies in raising gas price in this hard

bargain are also often above reasonable level; for example, it wants to charge from Ukraine

price above $300, which is not only unfeasible in the long run, but also does not correspond to

correct spatial pricing. That is why it is important to make some legal changes to eliminate

such games in future. There exists several possibilities for that (they may work apart, but can

be complementary): a) setting royalty and export tax for Gazprom monopoly at such level that

transit brings zero profit, b) building alternative pipelines; c) setting international law about

reasonable interval of transit prices based on audit of different firms.

20

7. Conclusions and Policy Implications

The present study tries to look beyond the year 2030. The analysis of the long term trends

shows that the demand for gas imports will grow very fast in the USA and Pacific Asia. This

would put Europe into fierce competition for gas supplies with other main gas consumers. The

diversification of Europe’s gas imports to mitigate its dependence on Russia looks favourable

but only for the next 1-2 decades. Beyond that Russia’s role as gas supplier will become even

more dominant facing only competitors from the Middle East. Summarizing, by the middle of

the 21st century most of the gas reserves on American and African continents will be depleted.

By that time Russia and Middle East would become the main gas suppliers. There will be

oligopoly-oligopsony relationship, with 3 main producers (Russia, Iran, Qatar) and 3 large

consumers (USA, EU, Pacific Asia). As a consequence, Europe will have to compete in the

future not only with Asia, but also with the USA. Therefore, Europe will, whether it likes it or

not, depend on Russian gas. The emergence of an Asian market with China up front, and the

possibility of LNG exports frees Russia from its dependence on gas exports to Western

Europe. On the other hand with Europe united and thus itself a powerful player and the huge

volume in question Russia has an economic incentive to portray itself as a reliable supplier in

order to increase the marketing prospects of its gas. In the long run, the only viable alternative

for gas supply in EU other than Russia is the Middle East. Therefore, EU should either invest

in pipelines to Iran, or be prepared to pay higher prices for LNG imports from Middle East.

The paper also addresses the issue of spatial pricing of gas, in particular for Russia. It is

recommended to observe real transport price of gas using accounting data, and set spatial

price variation in according to that. The question of natural rent for gas is also discussed.

Setting this rent at appropriate level along with tax on export allows for support of the present

price pattern (spatial differentiation) as regulated monopoly. Gas transit games and the

conditions for their elimination are also discussed.

References

1. Davies, P. Quantifying Energy. BP Statistical Review of World Energy 2006. –

London, 14 June 2006. (See http://www.bp.com/statisticalreview)

2. Luxbacher, R. (2004), The Growing Importance of the Natural Gas Industry, Oil

and Money Conference, London, October 26, 2004.

21

3. Bozon, I., Campbell, W., Vahlenkamp, T. (2007), Europe and Russia: Charting

Energy Alliance, The McKinsey Quarterly: The Online Journal of McKinsey &Co,

January 2007.

4. Kydland, F., Prescott, E. (1977), Rules rather than Discretion: The Inconsistency

of Optimal Plans, Journal of Political Economy 85, 1977, 473-493.

5. Golombek R., Hoel M., Vislie J. (editors) (1987), Natural Gas Markets and

Contracts, North Holland, 1987.

6. Monaghan, A., (2006), In: The EU-Russia Review. Issue 2. A report

commissioned by the EU-Russia Centre, November 2006, p.27-32.

7. Wirl, F. (1987), Joint Production of Substitutable, Exhaustible Resources or: Is

Flaring Gas Rational, Journal of Economic Dynamics and Control, 11, 495-511.

8. Stern, J. (2006), The New Security Environment for European Gas: Worsening

Geopolitics and Increasing Global Competition for LNG. – NG15, Oxford Institute

for Energy Studies.

9. Energy for a Changing World. An Energy Policy for Europe – the Need for

Action. European Commission, DG for Energy and Transport, Brussels, 2007.

10. World Oil Outlook 2007, OPEC, 142 p.

11. U.S. Geological Survey World Petrolium Assessment 2000.

12. World Energy Outlook 2006, IEA.

13. Yegorov Y., Wirl F. (2007), Energy Relations between Russia and EU with

Emphasis on Natural Gas, 6th Conference on Applied Infrastructure Research, TU

Berlin, Germany, 5-6 October 2007.

14. Chukmarova E. (2008) Za gazovij bunt Ukraina otvetit dolgom (in Russian);

(http://bigness.ru/articles/2008-01-18/gas_dolg/4668)

22