Pipeline Power

Franz Hubert Onur Cobanli

Humboldt University, Berlin Humboldt University, Berlin

hubert@wiwi.hu-berlin.de onur.cobanli@hu-berlin.de

Presentation for the ENERDAY 201227 April 2012

1 / 23

Motivation

The EU’s dependency on natural gas:

Russia (40% of imports, 25% of consumption)

The transit countries, Ukraine and Belarus (75% and 25% ofRussian imports, respectively)

New pipeline links (Nord Stream & South Stream) with Russia:

Diversify transit routes for Russian gas, but

Increase dependency on Russia and may

Reduce viability of investments in alternative sources(Nabucco)

2 / 23

Motivation

In 2008 Europe’s

Consumption: 489.7 bcm

Production: 184.2 bcm

Net imports: 305.5 bcm (Source: BP (2009), Statistical Review of World Energy)

Nord Stream and South Stream will increase transport capacity forRussian gas from app. 186 bcm/a to 304 bcm/a (63%).

All three pipelines together will increase the European pipelineimport capacity by 150 bcm/a (47%).

3 / 23

Motivation

Neither supply nor demand is there. The focus is on thestrategic role of the pipelines and the balance of power in thenetwork.

For the initiators of Nord Stream, Russia and Germany, thegains in bargaining power clearly justify the cost of investment.It is in the interest of the EU to support the project.

With Nord Stream already in place, the additional leverageobtained through South Stream is too small to make theproject viable.

The benefits through Nabucco accrue mainly to Turkey. Thegains for the EU, in contrast, are negligible.

4 / 23

Literature

Non-cooperative approach

Grais & Zheng (1996), Boots et al. (2004), von Hirschhausen etal. (2005) and Holz et al. (2008)

Computational advantages

Counterfactual assumptions from standard Cournot andBertrand set up instead of price-quantity contracts

Ad hoc assumptions on the nature of strategic interaction atthe various stages, the sequencing of actions and the ability tocommit

5 / 23

Literature

Cooperative approach

Hubert & Ikonnikova (2011a)

Efficient use of the existing network

Derive power structure endogenously from the actor’s role ingas production, transport and consumption

Narrow regional scope

Hubert & Orlova (2012)Mergers and the liberalizations of access rights within the EU

6 / 23

Model

NS

OS

NW

SW

West

Center

East

TC

AlgeriaLibya

Iberia

Balkan

Belarus

Belgium

France

Iran

Iraq

Italy

Poland

RussiaC

RussiaN

RussiaP

RussiaS

Turkey TurkeyE

UK

Center-

East

Center

Nether-

lands

UkraineKazakh-

stan

Turkmen-

istan

Uzbek-

istan

Azer-

baijan

Norway

NABUCCO

NO

RD

STREAM

SOUTHSTREAM

7 / 23

Model

Set of nodes R:

RP : Production

RT : Transit connections

RC : Customer

A link l = {i, j}, i 6= j ∈ R connects two nodes and has a capacitylimit kij and specific transportation costs Tij(x).

xij denotes gas flows from i to j.

8 / 23

Model

The value (or characteristic) function v : 2|N | → R+ gives themaximal payoff, which a subset of players S ⊆ N can achieve.

The value function captures the essential economics features, suchas the geography of the network, different cost of alternativepipelines, demand for gas in the different regions, production cost,ownership and access rights, etc.

For any coalition S ⊆ N we have to determine to which pipelinesL(S) ⊆ L the coalition S has access.

9 / 23

Model

v(S) = max{xij |{i,j}∈L(S)}

∑{i,j}∈L(S), j∈RC

∫ xij

0

pj(z)dz −∑

{i,j}∈L(S)

Tij(xij)

(1)

subject to

the node-balancing constraints∑

i xit =∑

j xtj , ∀ t ∈ RT (S)

the capacity constraints of the network |xij | ≤ kij , ∀ {i, j} ∈ L(S)

and non-negativity constraints xij ≥ 0, ∀ i ∈ RP or j ∈ RC .

10 / 23

Model

The Shapley value assigns a share of the surplus fromcooperation to each player.

φi, i ∈ N , which is player i’s weighted contribution to possiblecoalitions:

φi(v) =∑S:i/∈S

P (S) [v(S ∪ i)− v(S)] (2)

where P (S) = |S|! (|N | − |S| − 1)!/|N |! is the weight of coalitionS.

φi(v1)− φi(vo) : the gross impact of the pipeline on the surplus of

player i, which is then compared to the investment cost of thepipeline

11 / 23

Model

Access rights

Within the EU: Open third party access (TPA) to theinternational high pressure transport pipelines

Outside the EU: Every country has unrestricted control overits pipelines and gas fields.

Short horizon

A stationary environment with constant demand, technology,production cost, etc. All pipelines can be made bi-directional, butcapacities cannot be increased.

12 / 23

Calibration

Data for 2009 from IEA (2010a) on consumption and productionin the regions and flows between the regions

Constant production cost up to the production levels achieved in2009

Linear demand functions with the same intercept for all regions

Slope parameters estimated as to replicate the consumption in2009, given assumption on production cost

13 / 23

Calibration

The pipeline system as existing in 2009 is sufficient.

Given the willingness to pay and the cost of producing gas, it isable to deliver the efficient amount of gas into the differentconsumption nodes.

None of the expensive pipeline projects considered in this papercan be justified in narrow economic terms.

14 / 23

Calibration

The main differences between the power of the regions rely on:

Relation of total consumption to own production and not ondemand functions on which information is poor

Production capacity and pipeline connections to the marketsand not on differences in wellhead production cost which aredifficult to estimate

15 / 23

Results - Nord Stream

Nord Stream’s total strategic value for the initiators of theconsortium, Wintershall and EON Ruhrgas of Germany andGazprom of Russia (in our model Center and Russia) clearlyexceeds the project’s cost.

Transport competition mitigates the power of Ukraine and Belarus.

Norway and Netherlands suffer due to supply competition in theEuropean markets.

It is in the interest of the EU to support the project.

16 / 23

Table 1: Nord Stream’s Impact on Bargaining Power

Shapleyvalue [bn e/a]without with

Nord Stream Nord Stream differenceRussia 4.6 6.1 1.5

Ukraine 2. 0.6 −1.3Belarus 0.6 0.2 −0.4Norway 5.2 4.2 −1.

Netherlands 3.9 3.3 −0.6Center 9.4 10.1 0.7

Center-East 5.7 6.1 0.4Italy 1.7 1.9 0.2

Poland 0.9 1. 0.1France 4. 4.3 0.3

project costa 1.5aInvestment cost annualized with an interest of 15%.

17 / 23

Results - South Stream

Both South Stream and Nord Stream have almost identical effectson the power structure since both projects

bypass the transit countries

allow Russia to compete more effectively with Norway andNetherlands

protects Russia’s strong position in the Southeast

In the presence of Nord Stream’s large capacities, South Streamprovides much less additional leverage. The gains for theconsortium are not large enough to compensate for the project’shigh cost.

18 / 23

Table 2: South Stream’s Impact on Bargaining Power

with Nord Stream[bn e/a]Shapley Impact of pipeline sections

value (difference to column 3 table 1)OS OS+NW OS+NW+SW

Russia 6.5 0.2 0.4 0.4Ukraine 0.2 −0.2 −0.4 −0.4Belarus 0.1 0. −0.1 −0.1Norway 3.9 0. −0.2 −0.3

Netherlands 3.2 0. −0.1 −0.1Center 10.4 0. 0.2 0.2

Center-East 6.2 0. 0.1 0.1Italy 1.9 0. 0. 0.1

Balkan 0.5 0.1 0.1 0.1project costa 1.3 1.8 2.3

a Investment cost annualized with an interest of 15%.

19 / 23

Results - Nabucco

The whole Nabucco project just breaks even.

The lion’s share of the benefits accrues to Turkey and Iraq whilethe impact on the European regions is very small.

Supply competition harms Russia.

However, starting with the eastern parts, the incremental gains ofbargaining power do not cover the incremental cost of the centraland the western sections.

South Stream has almost no impact on the strategic viability ofNabucco.

20 / 23

Table 3: Nabucco’s Impact on Bargaining Power

without South Stream a with South Stream a

Shapley Impact of pipeline sections Shapley Impactb

value (difference to column 3 table 1) valueTC+ES WS TC+ES+ TC+ES+

CS+WS CS+WSRussia 4.7 −0.8 −0.3 −1.4 5. −1.5

Norway 4. −0.1 0.3 −0.1 3.8 −0.1Netherlands 3.2 0. 0.2 −0.1 3.1 0.

Center 10.2 0. −0.1 0.1 10.5 0.1Center-East 6.2 0. −0.1 0.1 6.3 0.1

Italy 1.9 0. 0. 0. 1.9 0.France 4.4 0. −0.1 0. 4.5 0.Balkan 0.6 0.1 0.1 0.1 0.6 0.1Turkey 5.1 0.7 0.5 1.3 5.1 1.2

Iraq 0.3 0.4 0. 0.3 0.3 0.3Iran 0.6 −0.1 −0.2 −0.1 0.5 0.

Azerbaijan 0.2 0. −0.1 0. 0.1 0.Turkmenistan 0.1 0. 0. 0.1 0.1 0.1project costc 0.9 0.5 1.7 1.7aAll figures in [bn e/a]bDifference to column 2 table 2cInvestment cost annualized with an interest of 15%.

21 / 23

Robustness

The relation of demand intercept and production cost determinesthe overall surplus from the gas trade.

With respect to an aggregate increase of demand in relation toproduction cost:

the relative shares of different players tend to be rather robust

the absolute values of their shares will increase, and as aresult more pipeline projects will become strategically viablefor given investment cost

Exclusive TPA within the EU: Conclusions regarding the strategicviability of the projects remain valid.

22 / 23

Thank you for your attention!

Questions?

23 / 23

Shapley Value - Example 1

N = {a, b, c}

v(a) = 0; v(b) = 0; v(c) = 0

v(a, b) = 0; v(a, c) = 0; v(b, c) = 0

v(a, b, c) = 1

Then, φa(v) = φb(v) = φc(v) = 1/3

24 / 23

Shapley Value - Example 2

N = {a, b, c}

v(a) = 0; v(b) = 0; v(c) = 0

v(a, b) = 1; v(a, c) = 0; v(b, c) = 0

v(a, b, c) = 1

Then, φa(v) = φb(v) = 1/2;φc(v) = 0

25 / 23