Water Diplomacy in Central Asia: Explaining the Lack of Cooperation

(1)

Water Diplomacy in Central Asia:

Explaining the Lack of Cooperation

Author: Julia Soldatiuk

Student number: 2251795

Supervisor: Dr. Francesco Giumelli

Supervisor Clingendael Institute: Dr. Louise van Schaik University of Groningen

(2)

(3)

Introduction

Water is of the great importance for Central Asia. The two main rivers, the Syrdarja and Amudarja provide water for the large agricultural sectors of Uzbekistan,

Turkmenistan and Kazakhstan, and for hydropower plants in Tajikistan and

Kyrgyzstan. The climate in Central Asia is moderate to dry, which is why agriculture in the region is not possible without extensive irrigation.1 When the Central Asian countries were still part of the Soviet Union, the area under irrigation was expanded, and sophisticated water distribution systems were created. In total 45 dams, 80 reservoirs and 20 000 miles of canals were build to satisfy the demand for water for irrigation.2_{Water in Central Asia was managed for a long time by one country: first} by the Russian Empire, and later by the Soviet Union. As a result, the countries were connected by a highly integrated economic system and were used to cooperate closely for many decades.3

The collapse of the Soviet Union brought an end to the joint economy and the joint management of water resources. Since each country was searching for a way to adapt its economy to the changed circumstances, water has become a matter of competition and sometimes even conflict.4 With the assistance of the international community,5 Central Asian countries have created a new mechanism to manage the water in the region. On 12 October 1991 the ministers of water resources of Kazakhstan,

Uzbekistan, Turkmenistan, Kyrgyzstan and Tajikistan signed an agreement with the telling name “On Cooperation in the Field of Joint Management and Conservation of Interstate Water Resources”6, thereby committing themselves to water cooperation in the Syrdarja and the Amudarja River Basins. This agreement, known as the Almaty Agreement, has established the Interstate Commission for Water Coordination that would coordinate transboundary river cooperation in the region and manage the implementation of the agreement. Central Asian states have also committed

themselves to “developing principles for fair regulation of water consumption taking

1_{B. Mosello, 2008: 152.} 2_{A. Khamzayeva, 2009: 11.} 3_{GIZ, 2013.}

4_{E. Weinthal, 2006.}

(5)

into consideration the interests of all nations of the region”7 and “to avoid unilateral actions that could adversely affect the neighboring republics.”8 In the years thereafter a number of other initiatives related to water cooperation were established,9 but like the Almaty agreement, they didn’t succeed in enhancing water cooperation.

Moreover, various reports10 revealed that the Almaty Agreement is frequently violated. States divert more water than was allocated to them, and Tajikistan and Kyrgyzstan increased hydropower production in the winter, leaving an insufficient amount of water for agricultural irrigation for the lower lying countries in the summer.

The aim of this thesis is to find out why water cooperation in Central Asia failed, despite the common past of the riparian states and the initially successful attempts to promote cooperation. The problem addressed in this research is the lack of

cooperation, meaning not the absence of cooperation as a whole, but rather the problems, obstacles and inefficiencies in the already existing cooperation on water between Central Asian states.

According to the definition of Robert Keohane, cooperation is characterised by mutual benefits and the presence of a goal that can be fulfilled through working together.11 This also means that if there are no mutual benefits, or if all goals are fulfilled, states might not feel the motivation to cooperate. However, if cooperation is not only beneficial, but seen as a key to the economic development of a region, as in the case of Central Asia, how can the lack of cooperation be explained in such a case?

The abovementioned considerations lead to the formulation of the research question of this thesis, which is as follows: How can the lack of water cooperation in Central Asia be explained?

To answer this question, an analysis will be made of the economic and political reasons behind the lack of cooperation. This thesis will first access the presence of

7_{V. Dukhovny, J. de Schutter, 2011: 217.} 8_{V. Dukhovny, J. de Schutter, 2011: 217.} 9_{A. Khamzayeva, 2009: 12.}

10_{International Crisis Group, 2014; K. Wegerich, 2008; L. Zakhirova, 2013; S. O’Hara, 2000; B.}

Mosello, 2010; E. Antipova, 2009.

(6)

economic motivations to cooperate, such as mutual benefits for states. Then the thesis will analyse what the political reasons are, and whether there are goals present that can be reached through cooperation. The theory of this thesis is based on the concepts of Game Theory, because it provides an overview of the factors that influence the decisions of states whether to engage in cooperation or not. Game Theory allows comparing states’ positions in cooperation, such as the power they have, their interests and the extent to which they can benefit from cooperation.12 While Game Theory uses Realists’ assumptions, such as that states are rational actors that are guided by self-interest, it is also often utilised by Liberals to explain the dynamics of interaction between the stakeholders in cooperation, and to determine the most

beneficial setting of cooperation for all participants.13 It has to be noted that this thesis will only analyse and make conclusions about the cooperation regarding

transboundary rivers in Central Asia. Therefore, this research does not say anything about the cooperation between Central Asian states in other fields.

The thesis will be structured as following. Chapter 1 will begin with a discussion on how the main schools of thought in International Relations, the Realist and the Liberal, explain cooperation. Then, economic and political approaches to water cooperation will be presented. Following, the Chapter will deal with Game Theory. Additionally, the concepts of Cooperative Game Theory that will be used to analyse the presence of the economic benefits will be presented. After that, three factors will be defined that will be used to analyse the lack of cooperation in the third chapter.

In the second Chapter the case study will be introduced, followed by an overview of the key aspects in water cooperation since the fall of the Soviet Union until now. Moreover, different possibilities of cooperation in bilateral and multilateral settings will be assessed using the principles of the Cooperative Game Theory. States will be grouped into coalitions in order to analyse the interests of each state and the extent to which each coalition fulfils it. The data and the setting of the coalitions will be based on the Almaty Agreement of 1992.

12_{A. Wolf, A. Dinar, 1997.}

(7)

In the third Chapter the factors that were determined in the first Chapter will be applied: the cost-benefit ratio of cooperation, the conflict of interests between upstream and downstream states, and power asymmetry between states. This chapter will consist out of four parts. The first part will look at the presence of mutual benefits for the states, the possibility of an equitable division of benefits between the states, and at the costs of cooperation. The calculations for the first two conditions will be presented in Annex 1 and 2. The second part will analyse the presence of conflicts of interest between the upstream and the downstream states. This will be achieved with help of the Political Accounting System, which includes three criteria that are scored for each state: their issue position, their power and the salience of the issue to the state. The application of these criteria will allow finding out whether the conflict of interests is critical enough to halt cooperation. The third part will look at the presence of partial coalitions and whether they can harm cooperation, and at whether or not the interests of the weaker states in the cooperation are being undermined. Finally, the last part will analyse the results of the analysis. In conclusion the findings of the research will be summarised.

(8)

Chapter 1. Theoretical Framework

In this chapter the term cooperation will be defined first. Second, it will discuss how different theories of international relations describe cooperation, and point out the strengths and weaknesses in their approaches. There is an extensive amount of literature written on cooperation, and it would take a separate paper to review it all. Therefore, in this review the author will provide an overview of the main approaches on how cooperation has been studied and explained in the literature on international relations. The second part of the literature review will focus specifically on water cooperation. Third, the central concepts of Game Theory will be discussed.

Additional attention will be given to Cooperative Game Theory and its approach to the analysis of cooperation. And finally, three hypotheses will be developed that will be further applied to the case-study in the Chapter 3.

1.1. Keohane: a Definition of Cooperation

Cooperation has become a focus of studies in international relations since the second half of the 20th century, when the European Union and international organisations, such as the United Nations, came into existence. Liberals along with realists,

contributed greatly to the literature on cooperation. Before moving to the discussion of why and how cooperation takes place, the term “cooperation” should be defined.

While there are many differences in how cooperation is approached in the literature, most IR researchers define cooperation in a similar way. This research will use the definition of cooperation formulated by Robert Keohane. He defines cooperation as a phenomenon that takes place “when actors adjust their behaviour to the actual or anticipated preferences of others, through a process of policy coordination.”14_{In his} understanding policy coordination is the adjustment of a state’s policies with the goal to minimise harm for other states.15 Keohane also argues that cooperation implies that countries work together to satisfy their national goals (in other words, behaving as rational actors),16 and that cooperation provides participants with mutual gains.17 These two factors distinguish cooperation from non-cooperative behaviour. Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non- Non-

14_{R. Keohane in: Cherles Lindblom, 1965: 227.} 15_{R. Keohane, 1984: 6 - 10.}

16_Ibid

(9)

cooperation can range from a conflict, when one actor tries to reduce the gains of the other on purpose; and unilateral actions, when one actor tries to maximise gains for itself, regardless of the consequences for other actors.18 To summarise, the three points that define cooperation are: mutual gains for all states involved, a presence of a goal that can be satisfied through working together, and the adjustment of a state’s policies.

1.2. Views on International Cooperation: Liberals vs. Realists

A heated discussion has developed through the years among researchers that

attempted to explain how cooperation is achieved. Realists argue that cooperation is difficult to achieve, because states exist in an international anarchic system. The basic assumption of realism is that states are rational actors that strive for survival in a world of anarchy. Thomas Hobbes, argued that it is not in the nature of people to cooperate: humans are egoistic and motivated by self-preservation and their own well-being.19_{According to this interpretation, states are not likely to have a high degree of} collective interests but they can engage in short-term cooperation as long as it allows them to achieve their goals.20

Realists also argue that cooperation in the long-term is not possible, because of uncertainty about whether the gains a state gets from cooperation in the present, are bigger than the costs from cooperation the state will pay in the future

(relative gains). Jean-Jacques Rousseau points out that in conditions of uncertainty, people are more likely to choose to pursue their own interest and gain less, instead of pursuing a collective interest that on the long-term will bring more.21 His famous example is that of a stag hunt, whereby five individuals have to make a decision whether to cooperate to catch a stag that will satisfy the needs of all five or to leave the group and catch a hare that will satisfy the needs of only one person (the one who catches it). While each individual realises that cooperating to catch a stag will bring him more meat, he can’t be certain that other individuals will also choose cooperation instead of acting according to their self-interest and catch a hare. Therefore the

18_{H. Milner, 1992: 468.}

19_{J. Steans, 2010: 66.} 20_Ibid

(10)

hunters are more likely to choose not to cooperate.22 According to Rousseau the same principle applies for states in deciding whether to work together or not.

Additionally, realists believe that the lack of trust between states creates more competition and conflict between them.23 John Herz called this situation a “security dilemma.” According to him, if one state increases its military potential for its own defence, other states assume (because states are concerned with their survival in a condition of uncertainty) that this is happening for offensive reasons, and they will increase their arms as well, which will raise security concerns for its own security in the first state, which will try to increase its military potential even more.24

The founder of the neo-realism, Kenneth N. Waltz, proposed a more optimistic view on cooperation. He recognised that cooperation is more likely to happen under the guidance of international institutions, or another way of ordering an initially “anarchic international system.”25_{He argued that the presence of a hegemon that takes a}

governing role, enables cooperation. Similarly, Charles Kindleberger argued in his Hegemonic Stability Theory that the international system is inclined to instability, which can only be avoided if one actor takes a leadership role.26 Neo-realists provide examples of British (19th and early 20th century) and American (second half of 20th century) dominance in the world economy. According to them, cooperation, enhanced by international organisations is also only possible if a hegemon takes the lead. The hegemon enables cooperation because its functions as an authority above the states, removing the obstacle of uncertainty. Joanne Gowa argues that a hegemon can promote cooperation by pressuring states to adjust their policies in order to achieve mutual benefits (which are not necessarily equal).27 According to this logic,

cooperation is not possible in the absence of a hegemon, and international organisations cannot function properly as well in this case.28

(11)

Even with the leadership embraced by a hegemon, some level of distrust is present in the neo-realist interpretation. Kenneth N. Waltz argued that state’s concerns about cooperation being more beneficial for others that for themselves, and about cooperation increasing their dependence on other states, pose challenges to international cooperation.29

Realists made two main contributions to the study of cooperation. Firstly, they explained the conditions of cooperation between states in the absence of a world governing authority (in other words, in a state of anarchy). The rational reasoning of actors is central here, as well as the level of certainty that costs will not overweight gains from cooperation. Secondly, realists analysed how cooperation is possible in a situation of unequal power distribution. They pointed out the importance of the leading role embraced by a hegemon, and its ability to bring a certain degree of order in the international system and to enable cooperation. However, because realists view states as rational actors, they fail to account for the role of common values and religions in the motivation of states to cooperate. Realists also argue that cooperation is not possible without a hegemon. They fail to provide a valid explanation for the multiple instances of cooperation since the WWII between states, without the interference of a hegemon, but based on the mutual interests and commitment.30

Liberals, on the opposite side, argue that states are inclined to cooperate, since they are interdependent and are not likely to reach their goals on their own.31 Neo-liberal institutionalists do not view states as unitary actors, and argue that states are

(12)

would increase interdependence between states and trust in each other. This,

according to him, will change the attitude of states from competitive to cooperative. Moreover, increasing interdependence is the key to long-lasting peace among states. The most prominent example of his idea is the European Union.34

Additionally, David Mitrany, the founder of the functionalist theory, stated that cooperation could be enhanced by the engagement of experts instead of politicians in fields such as economy, technology and science. He argued that such decentralisation of power would lead to effective cooperation between states and within international organisations.35 Richard Cobden argued that cooperation could be enhanced by free trade. According to him, the latter enables contact and communication among states that leads to more trust and friendly relations, which in return, leads to long-term cooperation.36_{Ernst Haas, on the other hand, emphasises the importance of political} elites for successful cooperation. He states that good connections between political elites are necessary for the establishment of cooperation. 37

Liberals do not view states as the main actors in international relations. Instead, international institutions take the central place and are the driving force of

cooperation.38 Moreover, neo-liberal institutionalists argue that there is a possibility of cooperation in the absence of a hegemon. Robert Keohane in his “After

(13)

Addressing the realists point that institutions only exist as long as there is a hegemon, liberals argue that institutions can evolve over time and become independent actors in international politics that are not dependent on the shifts of power in states – therefore they can enhance long-term cooperation.41 Thus, institutions can indeed exist without the presence of a hegemon.

A number of scholars have focused on the question of how cooperation comes about. According to Helen Milter, there are three types of cooperation. First, there is the kind of cooperation that is imposed by a hegemonic power. Second, she mentions the kind of cooperation that emerges spontaneously, when actors naturally come to similar expectations and realise that the best way to achieve their goals is through

cooperation. Robert M. Axelrod is a supporter of this theory. He argues that no specific agreement is needed in order to foster cooperation.42_{And third, the kind of} cooperation that is the result of an extended negotiation process, where mutual gains and policy coordination are agreed upon.43 This way of thinking is supported by Kenneth Oye, Robert Keohane and Joseph M. Grieco, who claim that cooperation is achieved by a bargaining process, where the redistribution of gains plays a central role. According to Helen Milter, the second type of cooperation is more common. She notes, however, that negotiated cooperation has received more attention in IR, as it is easier to identify cooperation that is based on negotiations (as they are often

documented and have a starting date) than the cooperation that arises spontaneously.44 At the same time spontaneous cooperation develops into a negotiated cooperation with signing of an agreement that makes cooperation official.

All in all, liberals view the interdependency of states and the decentralisation of authority as important conditions for cooperation. States should also see mutual economic and political benefits and a long-term commitment of their partners in order to develop cooperation. International institutions play an important role in cooperation by reducing the level of uncertainty in international politics and creating a

rule-governed environment. While cooperation can be spontaneous, more often extended

41_{J. Steans, 2010: 42-43.}

42_{R. Axelrod, 1981: 1390 - 1396.} 43_{H. Milner, 1992: 469.}

(14)

negotiations that result in a treaty or another kind of binding document, precede the beginning of cooperation. According to liberal interpretation, cooperation between states is possible in the absence of a hegemon, and is in many cases desirable,

however still uncertain. While realists point out the conditions in which cooperation is not possible, liberalists do not provide clear arguments on this matter.

1.3. Views on International Water Cooperation

Previously it was discussed how cooperation is explained in international relations literature. Similar views are reflected in the research on international water

cooperation, which has developed into a separate branch in the last decades.45 This section will provide an overview of the different perspectives on water cooperation. But first, a definition will be given of the central concepts in water cooperation, such as freshwater cooperation and maritime cooperation, trans boundary river basins, riparian states, and upstream and downstream states.

Water cooperation can be defined as continuing interaction between two or more states or other actors with the goal of jointly approaching water resources to create benefits for each party involved.46 Water is an increasingly scarce resource that plays a key role in the development of each country. Water resources form an extended complex of rivers, lakes and underground waters that cannot be managed for one purpose or by one country. Usually water cooperation combines multiple purposes, from domestic and tourism to industrial and agricultural use.47 As was stated before, the goal of water cooperation is to create added benefits (that is benefits above what actors receive if they don’t cooperate). Those benefits include increased productivity in different industries, such as hydropower or agriculture, which gives access to new markets and the possibility to develop sectors of the economy that are water intensive. Also, water cooperation allows to diminish water losses by more efficient water use, improved management and the location of water infrastructure, as well as lowered

(15)

expenses on threats, such as floods, water pollution and climate change, by addressing them together.48

Depending on the origin of water, water cooperation can be divided in maritime and

freshwater cooperation. Maritime cooperation deals mostly with reaching agreements over disputed areas on sea, and is often approached from a legal perspective.49

Freshwater cooperation deals with cooperation in river basins, lakes and underground water resources. It is directed at creating an efficient allocation of water resources (to increase mutual benefits) and establishing agreements to prevent or solve conflicts over access to water.50 Cooperative Game Theory plays an important role in the research on freshwater cooperation, as it can provide schemes of possibilities for cooperation that are often used as a basis for negotiations.51 One of the central concepts in freshwater cooperation is a transboundary river basin.

Transboundary river basins constitute an area drained by the river and its tributaries that is situated in more than one country. These countries are called the riparian states. The drainage area can spread further than the river itself. In this case the country that only includes drainage area will still be classified as a riparian state.52 Riparian states are divided in upstream and downstream countries. Upstream countries are one or more countries where the river begins. These countries are usually more mountainous than downstream countries, and often they are responsible for the biggest share of the water that is flowing through the river. This, depending on circumstances, gives them the power to control the amount of water downstream countries will receive. Considerable hydropower potential is often a characteristic feature of upstream states.53 Downstream countries are situated in the middle and the end of the river flow and are often the main users of the water resources. In many cases downstream countries are more economically developed, compared to upstream states. While downstream states can also have hydropower potential, usually their

48_{A. Subramanian, B. Brown, A. Wolf, 2012: 2.} 49_{R. D. Hayton, 1990: 2, 45.}

50_{J. Karreth, K. Tir, 2013: 3.}

51_{A. Subramanian, B. Brown and A. Wolf, 2012: 2.} 52_{J. R. E., S. Netanyahu eds., 1997: 11.}

(16)

main interest is water for agriculture.54 Differing interests between upstream and downstream states can create tensions, lowering the possibility of cooperation.

When it comes to explaining international freshwater cooperation, some authors argue that economic factors play a key role in the success or failure of cooperation; others argue that political factors can provide the most valid explanation. This research will combine both perspectives, including economical and political factors in the analysis of water cooperation.

1.3.1. Economic perspective on water cooperation

Supporters of the economic perspective argue that the distribution of economic gains is central in enhancing cooperation.55 Their reasoning is built around two concepts: equitable water allocation and costs of cooperation. The first is related to the fair division of water shares between the riparian states, and the second is about the costs that have to be made to enhance cooperation. These include the costs of implementing the agreements and adjusting national policies, investments in and renovation of the river basin infrastructure (such as dams, and water monitoring systems) and other related costs.56

Aaron T. Wolf argued that cooperation often fails because states cannot manage to divide costs and benefits from the use of water in a fair and efficient way. He focused on researching the criteria for equitable water allocations and came up with two methods of water allocation. He emphasised the “allocation of water resources according to their economic value”,57 and argued that the latter should be efficient (water should be divided according to the value riparians attach to it) and equitable (gains from cooperation should be distributed according to the contribution of each participant).58 The distribution of water resources can be done using two approaches. The first one, called a “social planner”, views the river basin as a whole, and assumes that there is a central authority that choses the most efficient way to distribute the water resources and gains from cooperation between all participants in the basin. This

54_{J. R. E., S. Netanyahu eds., 1997.}

55_{P. H. Gleick, 1993; I. V. Goslin, 1977;J. V. Krutilla, 1969; K. E. Hayness, D. Whittington, 1981.} 56_{J. Bendor, D. Mookherjee, 1987: 129.}

57_{A. T. Wolf, 1999: 7.}

(17)

approach was used to analyse the Colorado river’s allocations between the US and Mexico,59 The Nile river’s allocations for its riparian states,60 as well as the benefits from cooperation between Turkey, Syria and Iraq in the Euphrates and Tigris River basins.61 The second, called a “water market” approach, views river basins as divided between the competing states, and assumes that states will divide the water resources through a market mechanism.62 This approach was applied to the division of water for various river basins in the Middle East.63 Both approaches have two conditions that determine whether equitable water divisions can be achieved. First, there have to be mutual benefits for all participants, and all costs and benefits from cooperation should be divided in a way that each participant receives more from cooperating than from not cooperating. Second, the division of costs and benefits should be done in a Grand Coalition that includes all participants; partial coalitions should be avoided.64

Returning to the methods of allocation of economic costs and benefits, Wolf argued that this could be done using the principles of “beneficial users” or of “baskets of benefits”.65 The principle of beneficial users employs that water should be allocated according to “its most efficient use”66, meaning that water should not be divided equally, but on the benefits from its use, based on the values each state attaches to water. This ensures satisfaction of the economic needs of each riparian state. This method of allocation includes a compensation mechanism: for unused potential in hydropower production by an upstream state, or for flooded land in a downstream state,67 which forms a stable basis for integral river basin use, and therefore, a long-lasting cooperation. The weakness of this method is that allocations based on values of water are difficult to change, because of the integral character of such agreements. They then have to be renegotiated, which is a lasting and expensive process.68

(18)

The principle of baskets of benefits is an issue-linkage method, where water negotiations are combined with other issues, such as financial, energy resources, political linkages, or data sharing (that allows other riparian states to see how the water is used in another country and propose a more efficient way of use).69 By receiving a larger water share, a country will make the concessions in other political or economical issues. This method is often used in practice. One of the examples is the case of Turkish - Syrian cooperation in the Euphrates river basin. Turkey, an upstream country, agreed to release more water for irrigation in Syria, and Syria in return, withdrew its support for Kurdish nationalists that were active in Turkey.70 This method has worked to enhance cooperation between the conflicting countries. For example, issue-linkage has enabled water cooperation in the Jordan River basin between Israel and Palestine. This is the only issue the two countries cooperate on.71 Some researchers argue that water cooperation between conflicting countries can have a spill over effect, and could enhance cooperation on other issues as well.72 This claim, however, has not yet received sufficient confirmation in practice, so whether this is really the case, remains to be seen.

It should be kept in mind that literature that evaluates economic benefits from water cooperation does not reflect the division of benefits from cooperation in reality. It merely proposes a scenario for the optimal allocation of water shares and the division of economic benefits that can be used as a basis for negotiations. The ideas listed above correspond with Keohane’s argument about mutual benefits being the driver of cooperation in the absence of a hegemon.

1.3.2. Political perspective on water cooperation

Economically equitable water solutions, however, are not always enough to foster cooperation.73 David LeMarquand argued that the political priorities of the states

69_{A. T. Wolf, 1999: 15.}

70_{P. H. Gleick, 1993: 88 - 89.}

71_{Water Diplomacy Consortium, 2014.} 72_{Water Diplomacy Consortium, 2014.} 73

A. T. Wolf recognized the importance of political considerations in water cooperation as well, and in his later research he made an attempt to combine the two approaches (inspired on the work of

(19)

could have a decisive influence on the course of cooperation as well. He distinguished a number of political factors that influence water cooperation.

Those are hydro-economic incentives, the upstream-downstream conflict, questions of national image, the linkage of river issues with other foreign policy issues,

reciprocities that each nation may have, and the general climate for or against agreement.74

Regarding hydro-economic incentives, states tend to see the river as their property, thereby ignoring the fact that other countries also make use of it. Neglecting the negative consequences for the other states (such as environmental degradation, floods or shortages of water), caused by the overuse of the river, often leads to interstate tensions and conflicts. According to LeMarquand, viewing rivers as natural resources that belong to all riparian states would promote cooperation.

LeMarquard builds up his reasoning using the concept of public goods, which implies that every country in the basin has equal access to the river, and can use the river according to its needs without diminishing the benefits of its use for other countries. Another key concept is that of common pool resources, according to which each riparian country is entitled to use the river according to its needs, but the benefits from its use should be proportionate to the costs it pays to keep the river in the good condition.75 This is because the common pool resources concept implies that intensive use by one state diminished the possibility to profit from the river for another state. LeMarquand argues that giving a river the status of public good allows equal access to the river for navigation, increases trade and other types of exchanges, and

therefore, promotes cooperation. Viewing water as a common pool resource, makes cooperation much more likely, since the riparian states recognise that the country’s use of water and the costs a country bears to prevent harm to the river have to be in proportion. Furthermore, a joint agreement of riparian states to integrated river development provides more benefits (such as flood control projects, dam

construction, compensation for lost hydro energy generation).76 The latter creates an additional motivation for riparian states to cooperate.

(20)

Regarding the upstream-downstream conflict, LeMarquand explains that the

difference in interests between the upstream and the downstream states often hinders cooperation. Conflict arises when upstream countries use their situation to limit access to water for other riparian states by building dams and reservoirs. By releasing water seasonally they can cause floods in downstream countries.77 A state that

controls the source of the river can have a stronger position in negotiations, or even be a key player that determines whether agreement will be reached or not.78 This is the case in Yarmouk River Basin, where Syria, as an upstream country has a dominant position in negotiations with Jordan.79 However, if the downstream state is more powerful, it can lead negotiations. A stalemate is also a possibility in such a situation, especially if downstream and upstream states have opposing claims. This is the case in the Nile Basin, where Egypt, as a downstream state, claims a water share of more than half the entire river flow, and agrees to negotiate only if the country can keep a leading role in the negotiation process. Ethiopia, being an upstream state and having the strongest military power in the region, also wants to receive a substantial water share (previous water agreement between Egypt and Sudan, excluded Ethiopia and nine other states). This situation has resulted in frozen negotiations and dangerously growing tensions in the river basin.80

Regarding the third factor, the national image of the country, LeMarquand argues that the willingness of a country to appear as a responsible actor or neighbour, or to influence the public opinion in another country, might stimulate the state to

cooperate.81 He illustrates his point with the example of water cooperation between the United States and Mexico in the Colorado River Basin. While the terms of the agreement were not economically beneficial for the United States, the state cooperated to create a positive image of itself in Mexico and the rest of Latin America, in order to improve its position in the region.82

(21)

Regarding linkage to other foreign policy issues, LeMarquand’s argument

corresponds with Wolfs “basket of benefits” approach. The difference here is that a state’s motivation to cooperate comes not only from getting concessions on issues such as trade agreements or support for a multilateral policy, but also from creating a “reservoir of a good will”83 that can be used later for support on another issue.

Cooperation is also more likely if states have a feeling of a mutual commitment or mutual obligations, often based on a similar ideology (such as the commitment to promote democracy, defend human rights, or to prevent climate change). In this case the motivation of states to cooperate will have an economic or psychological

character.

And finally, regarding the climate for an agreement: the more similar views riparian countries have on the issue, the higher the probability of the water cooperation. Here, different sources suggest different preconditions for water cooperation. Scott argues that a similar vision on environmental quality or river basin development, as well as well-established contacts between political elites, promote cooperation in a river basin;84 Russell Sullivan emphasise that cooperation is more likely with a small number of riparian countries, which make the costs and benefits for each state more significant; if one country produces goods that could benefit other riparians; and when there is a presence of a large riparian state, that is prepared to take responsibility for the additional costs from cooperation. 85

(22)

Accounting System (PAS) to the Middle East water-trading game, where they analyzed the behavior of states influenced by political pressures in their countries. Based on their calculations, the research determines the probability of cooperation taking place.88 The PAS method was introduced by Coplin and O’Leary in 1974. The method analyses three dimensions: the issue position, the power of a player to

influence an issue, and salience. The first criterion indicates the strength of each player’s position, whether he supports or opposes the coalition scenario. The second criterion looks at the level of influence which each of the actors can exercise on others and the outcome of the game. And the last criterion expresses the importance of the negotiated issue to each player.89

T. Naff and R. C. Matson proposed another approach to calculate the political likelihood of water cooperation between riparian states. They used three criteria: the motivation of players to participate, the riparian position regarding the water, and the power to prevent any coalitional agreement. This method in his essence is very similar to PAS, but its interpretation is different. In PAS conclusions are drawn based on the comparison of more than one scenario. Comparisons are made based on the total outcome of coalitions. The lower the total outcome, the lower the possibility of cooperation.90 Naff’s and Matson’s approach compares the total outcome for each player in a coalition. The more equal players are ranked, the higher potential for conflict.91

To summarise, the economic perspective on water cooperation points out the importance of an equitable division of all costs and benefits from cooperation. It emphasises that water should be divided according to its value to riparian states, and that including other issues and resources to negotiations on water cooperation could enhance chances on cooperation. From a political perspective, there is a number of factors that influence water cooperation. The way states perceive a river basin and themselves, influences their decisions and their willingness to cooperate. If riparian countries share a vision on the use of the river basin and have well-established communication between them, than water cooperation will also be more likely.

88_{A. Dinar, A. Wolf, 1997: 21.}

89_{W. Coplin, M. O’Leary,1974; A. Dinar, A. Wolf, 1997: 21.} 90_{A. Dinar, A.Wolf, 1997: 9.}

(23)

Interestingly, some features of the economic approach correspond with the realists views on cooperation, while some of the arguments made by supporters of the political approach correspond with the liberal thoughts. For instance, the idea of a “social planner” approach corresponds with the argument that cooperation is only possible if a hegemon takes a leading role and creates a degree of certainty in an anarchic system. The social planner assumes a presence of a central authority that would take responsibility for equitable water allocation in the river basin as well as for designing the most efficient way of water use. Liberals underline the

interdependence of states and argue that well-established communication between states is key to understanding each other’s interests and approaches. Supporters of a political perspective on water cooperation also point out the interdependence of the riparian states, and emphasize the importance of similar views and tastes for the river basin development for fostering water cooperation.

Despite the differences in how realists and liberals view cooperation, both camps agree that mutual gains play an important role in a state’s motivation to cooperate. Also, to study cooperation and non-cooperation, both camps have utilised Game Theory, which will be used in this thesis to explain the lack of water cooperation between Central Asian states. Both realists and liberals view states as rational actors, but realists usually use zero-sum game to analyse the behaviour of states in situation of anarchy, such as arms race or deterrence.92 They also turn to Game Theory to justify the claim that cooperation is not likely to arise in international relations. The already mentioned security dilemma of John Herz’s and stag hunt game of Jean-Jacques Rousseau’s, are among examples of the application of Game Theory. Liberals often use positive-sum game to explain why states do or do not cooperate.93 Liberals believe that the total profit can be increased if states can find a compromise and adjust their behaviour with relation to each other.

Game Theory is also often applied to research in water cooperation. In many cases the goal is to determine the most beneficial scenario of cooperation for all involved states. Game Theory in such cases allows to calculate the economic benefits of cooperation

(24)

and to determine the conditions that would be acceptable to all states to develop cooperation. Aaron Wolf and Ariel Dinar, the leading researchers in the field of water cooperation, used Cooperative Game Theory in their research to illustrate how self-oriented states can cooperate based on mutual benefits.94 The following section provides an outline of the key concepts of Game Theory and offers an outline of the framework of analysis.

1.4. Game Theory

John von Neumann and Oskar Morgenstern first introduced Game Theory in 1944.95 It was primarily used in mathematics and economics.96 Due to the broad variety of methods to calculate the behaviour of “players” in various situations, Game Theory was later used in many other fields. Anthony Dows was one of the first scientists that applied Game Theory to political science to analyse how the ideological distribution of voters in a community influences electoral outcomes.97_{Nowadays, the application} of Game Theory to political science and international relations ranges from

calculating the outcome of conflict situations and wars to the fair division of natural resources and public choice.

Game Theory is defined by Myerson as “the study of mathematical models of conflict and cooperation between intelligent rational decision-makers”.98 Game Theory studies interactions among players that make decisions to secure the best outcomes for

themselves in the condition of uncertainty.99 Interdependence is a key feature in every game.100Players are defined as stakeholders that interact with each other in a game. Players are the ones who decide whether to cooperate. Players are often states, but they can be also individuals, governments, companies, municipalities; in other words, players are those who’s behavior and interaction is being analyzed.101

Depending on the outcomes for each of the players, Game Theory is divided on zero-sum, positive-sum and negative-sum games. In zero-sum the interests of all players

94_{A. Wolf, A. Dinar, 1992, 1994, 1997.} 95_{J. von Neumann, O. Morgenstern, 1944.}

96_{D. Avinash, B. Nalebuff, 2008; D. Avinash, B. Nalebuff, 1991.} 97_{A. Downs, 1957: 135 -150.}

(25)

are conflicting, which means that if one player wins, another will inevitably loose.

Positive-sum game means that mutual gains for multiple players are possible, although not necessarily equal. In negative-sum game both players will loose, no matter which choice they make.102

Depending on the focus of the study, Game Theory can be divided into two main branches: non-cooperative and cooperative games. While cooperative Game Theory focuses on values, non-cooperative game focuses on the moves and strategies of each individual player in order to predict which actions they will make, and how they will influence the payoffs.103 In non-cooperative theory players don’t know the moves of their opponent in advance, while in cooperative game players are aware of the possible outcomes for them and other players.104 Non-cooperative game is suitable for analysing elections, wars or for resolving issues between conflicting parties, where rules of the game are clear and precise.105_{It is usually a zero-sum game, meaning that} gains of one player are loses of another.

Cooperative game is a game where players can form a coalition in order to maximize their benefits, assuming that their payoffs will be bigger in cooperation than in competition.106 Cooperative game calculates the added value the players will attain forming a coalition and tries to predict which coalition will be formed and which payoffs players may obtain.107 Cooperative game is usually a positive-sum game that can be applied to negotiation process, formation of government or allies between countries, but also to analyse the market interactions, where various players and interests are present.108 Cooperative game is also used for the division of limited natural resources, where there is a focus on maximising benefits for players by cooperation, taking into the account the necessity to share limited resources.109

102_{O. Chataim, 2014: 1.}

103_{O. Chataim, 2014: 1.} 104_{R. L. Teasley, 2009: 14.}

105_{Interview with Robert Aumann in: L. B. Martinez, 2009: 2 - 3.} 106_{O. Chataim, 2014: 1.}

107_{O. Chataim, 2014: 1.}

(26)

This research will focus on the Cooperative Game Theory because it aims to reveal the reasons behind lack of cooperation on water between Central Asian states. As was mentioned before, the goal of cooperation is for each state to achieve more benefits. This implies mutual gains, and this is best approached as a positive-sum game.

1.5. Cooperative Game Theory

Cooperative Game Theory usually has more than two players. Groups of players can communicate before and during the course of the game, and are expected to make agreements that are called Coalitions. Coalitions can be formed by all players in the game (grand coalitions), by few players (partial coalitions), and by one player in the game (then they are called non-cooperative).110 Competition here is present between coalitions, rather than between individual players (as in non-cooperative game). Players in a game can represent all stakeholders, such as states, industries, individual resource users and regional institutions. To calculate the collective benefits players will get, forming a coalition, a Characteristic Function is used. It shows the total benefit countries will get in a coalition, by calculating a value or a profit to each coalition in a game.111 The Characteristic function for a transboundary river basin cooperation game in this research has the following expression:112

𝑀𝑎𝑥 𝑍 = − 𝑤_!" 𝐸!"#$%! !" _{− 𝐸} !!"#$!" 𝐸_!"#$%!!" )!+ 𝑤!" 𝑃_!"𝐴𝑟𝑒𝑎_!" 𝑀𝑎𝑥_!"#$%&!" ! !!! + 𝑤_!"#$𝑃!"#$𝐴𝑟𝑒𝑎!"#$ 𝑀𝑎𝑥_!"#$%&!"#$ + 𝑤!"#$ 𝑃_!"#$𝐴𝑟𝑒𝑎_!"#$ 𝑀𝑎𝑥_!"#$%&!"#$ Where: 𝑀𝑎𝑥 𝑍 – is the Characteristic function for all players, 𝐸_!"#$%!!" – is the annual hydropower demand of an upstream state (gWh), 𝐸_!!"!"!" – is the annual hydropower production of an upstream state (gWh), 𝑃_!" , 𝑃_!"#$- is the agricultural net margin ($/ha) of an upstream or a downstream state, 𝐴𝑟𝑒𝑎_!", 𝐴𝑟𝑒𝑎_!"#$ – is the agricultural area (ha) of an upstream or a downstream state,

𝑀𝑎𝑥_!"#$%&!", 𝑀𝑎𝑥_!"#$%&!"#$ – is the maximum possible annual profit ($), 𝑤_!", 𝑤_!"#$- are the nonnegative weights, they represent the relative importance of

110_{R. L. Teasley, 2009: 14.} 111_{J. Drechsel, 2010: 10.}

(27)

satisfying the various objectives of the countries. They have values less than one and a collective sum of one.113

When all coalitions and their characteristic values are determined, allocation methods are used to divide payoffs of each player in a coalition. The basic assumption is that each player should receive a share not smaller than what he would receive without cooperation. The combination of possible allocations of benefits within a coalition, to which each player will agree, is called the Core of the game. The core provides the range of minimum and maximum benefits a player would get from cooperation. The bigger the range, the more space players have for negotiation.114 If the core does not exist, the interaction and cooperation between players will not be stable.115 The core has two values: the lower and upper bound. The lower bound is the lowest value a country can get, the upper bound is the highest value in a coalition minus the value of a coalition without the country for which the core is being calculated. 116

When the core is established, the Shapley allocation117 is used to distribute the benefits from cooperation in a coalition between all players, according to their contribution. The Shapley allocation, which was introduced by Nobel Price winner Lloyd Shapley in 1953, shows how much each player contributes to the creation of added benefits in the coalition. Since the contribution to the coalition and the bargaining power (the relative ability of players to exercise influence over each other)118 may differ among the players, the payoff they get will also not be equal. Therefore an allocation method, such as Shapley is needed to calculate the outcomes for each player. The Shapley value (𝜃_!) in this study is expressed as:

𝜃_! = 𝑛 − 𝑠 ! 𝑠 − 1 !

𝑛! (𝑣 𝑠 − 𝑣(𝑠/{𝑗}))

Where n is the number of players in a game and |s| is the number of players in

coalition s. 𝑣 𝑠 – payoff for coalition s, 𝑣(𝑠/{𝑗})– is the payoff for coalition s without player j. For a better overview the Shapley value is converted into percentages. The Shapley allocation also represents the added value of each player to a coalition, or as 113_{R. L. Teasley, 2009: 128.} 114_{R. L. Teasley, 2009: 14.} 115_{O. Chataim, 2014: 2.} 116_{J. Drechsel, 2010: 18 – 36.} 117_{L. S. Shapley, 1953.}

(28)

O. Chataim puts it, “the value lost to the grand coalition if the agent is not included.”119

Cooperative games often include the calculation of a coalition stability. It shows the likelihood of one or more players leaving the coalition because they are not satisfied with the share that is allocated to them according to the Shapley method. Coalition stability in this research is replaced with the political likelihood of cooperation (calculated with the PAS method), which is more extensive and better fitted for answering the research question.

As with every model, Game Theory models simplify the reality to reveal certain patterns in the interaction between the stakeholders. As a result, the application of Game Theory has a number of shortcomings that should be kept in mind while reading this research.

First, in order to analyse a specific aspect of a conflicting situation, the game neglects some factors while prioritising others. The information that is left out in the game can lead to misperception of the situation.120 For instance, a game can include states as a players but exclude multinational companies, municipalities, non-governmental organisations or other actors that also have an influence on the issue that is being analysed is a game. Game Theory experiments set certain parameters of the game, within which player is supposed to act, making rational choices. The goal of such a game is to collect information about the behaviour of players in a certain situation. In a game, players assume that they act in a one-shot game.

Second, the game doesn’t account for the possibility of future interactions between players. When players know they might deal with the same players in the future, they might choose a different strategy, often more cooperative. In this sense the Game Theory experiment will not reflect the actual behaviour of players if the situation in a game will occur in reality. 121

119_{O. Chataim, 2014: 2.}

120_{S. Brams in M. Brecher and Frank Harvey eds., 2002: 90.}

(29)

Third, Game Theory assumes that players have an overview of their pay-offs and the pay-offs of others, while in reality it is often not possible to get precise information about pay-offs and the possible outcomes of the game prior to the start.122 Despite its limitations, Game Theory still provides a valid tool for explaining the phenomenon of bargaining and the formation of coalitions. Also, it outlines the best option for

cooperation that can be used to define the most beneficial strategy. Based on

presented discussion of cooperation and the concepts of the Game Theory a number of preconditions of cooperation can be defined. They will be discussed in the following part of the chapter.

1.6. Analysing water cooperation

1.6.1. Cost-Benefit Ratio

The basic assumption of Game Theory is that players are rational beings and will act according to their own interest. This means that the decisions of states are influenced by their motivation to increase benefits they receive. From this follows that

cooperation is only possible if all players will receive mutual gains, or in other words, when all players will receive bigger profits in cooperation than if they would be acting alone. Related to this, is the concept of equal value, which holds that cooperation is possible if all players attach equal or similar value to the gains they will receive from cooperation.123 Benefits have a “transferable utility”124 meaning that the pay-offs of one player will have a similar value to those of other players. Pay-offs are additional gains of a player that he will only receive if he is engaging in

cooperation. Pay-offs are usually expressed in economic terms and, as money is valued by all states in a similar way, this condition of cooperation is usually met in cost-related games.

Players are not only concerned with absolute gains (total benefits they will receive from cooperation), but they also tend to focus on relative gains, comparing their benefits to the benefits of the other states.125 Every state is willing to receive more benefits than the other. These concerns over relative gains between states, according

122_{S. Brams in M. Brecher and Frank Harvey eds., 2002: 90.} 123_{H. Milner, 1992: 472.}

(30)

to game theorists, stimulate competition and diminish the possibility of cooperation. The solution to this problem provided by Game Theory is an equal distribution of absolute gains between the players (which is explained in the section 1.5.).126 Equal distribution does not mean that each state should receive the same pay-offs, but rather that the benefits the state receives would correspond with the contribution of the state to the coalition.127 This is called an equitable distribution of gains,128 and it can legitimise why players in the game receive different amounts. The Shapley value is used to calculate the contribution and the pay-offs for each state.

The costs of conducting cooperation also influence the willingness of players to cooperate. There are two types of costs: costs of cooperation itself (transaction, maintenance costs, costs of installation of water use monitoring equipment) and the costs of policy adjustment and implementation.129 The latter can be very high, especially if cooperation requires restructuring of certain sectors of the economy or adjusting administration. Costs should not overweight the benefits from

cooperation.130_{When the costs are higher than the benefits, cooperation is very} unlikely to occur. When the benefits are higher than the costs, the rule of thumb applies that the higher the difference between the benefits and the costs from

cooperation, the higher the chance that cooperation will occur. What should be noted, however, is that in practice not all costs as well as the benefits are evident/known before cooperation occurs. That is why decision-makers always have to take a risk and hope that their estimations will not deflect too much in the course of cooperation.

(31)

1.6.2. Conflict of interests

As was mentioned before, upstream and downstream states may have conflicting interests due to their geo-political position in the basin and their geographical and climate conditions. These differences can make it harder for the states to come to an agreement on cooperation. To analyse whether these conflicting interests between states have a decisive influence on cooperation, A. Dinar and A. Wolf developed a method, called the Political Accounting System (PAS), to analyze how political aspects influence cooperation. The central assumption here is that there are three factors that influence cooperation: issue position, power and salience. 131 Firstly, a proposed scenario should meet the needs and interests of a player. If it does, the player will support a proposed scenario, if it doesn’t, the player will oppose a proposed scenario. Secondly, every player possesses relative power to influence the decision and behaviour of other players. The level of power is determined by the relative military or financial power, the legal and bargaining position, and the level of trust in relation to other players. The higher the relative power the higher the

possibility that the outcome of the scenario will reflect the interests of this player. Thirdly, the issue being negotiated can be less or more important to the player.

The optimal scenario for cooperation is when all players agree (to a differing extend) with the scenario. In that case the power position and the importance of the issue become less significant. If all players oppose the scenario, cooperation very unlikely to happen. But if some players support and some oppose the scenario, their power position and the importance of the issues to them becomes decisive.132_{In this case,} players that have more power to influence others will determine the outcome of cooperation. The most difficult situation arises if both, players that oppose and that support the scenario have a strong power position. Cooperation then is highly difficult and such a scenario can lead to a stalemate or even a conflict.133

(32)

The factors in this game cannot be measured by a quantitative method. Arial Dinar and Aaron Wolf134 propose to determine the position of each player on a certain issue doing literature analysis that accounts relative military power of a state, legal standing and bargaining position (on water) within the region, as well as the level of hostility and trust between players. Another option is to set up a panel of experts on the issue and ask them to mark the factors for each player.135 Both methods will still provide a subjective definition of the player’s position. The second method requires time and financial expenses that are above the scope of this research. Therefore, all three factors are determined with the help of literature research.

The Issue Position is marked in the range between -3 and +3, where a lower value shows a negative and a higher value shows a positive attitude towards a certain proposal of cooperation (see Table 1). Power is ranked from 0 to 3 and is reflecting the relative power position between the riparian states, military and political power, as well as legal strength in the water-sharing agreement. Salience is marked from 0 to 3 and is reflecting the importance of the issue being negotiated.136 The marking is based on the relative position of one state to the others. To determine the total effect each player has on a joint decision, the three factors are multiplied. The total strength of a coalition is the sum of total effects of all players.137 Though too general, PAS

indicates stakeholders and processes that characterise cooperation and sheds light on the misperceptions players have of each other and the coalition.138

Issue Position

-3 Strongly opposes the coalition -2 Opposes the coalition

-1 Sees benefits but opposes the coalition 0 Neutral or issue is not relevant

1 Considers cooperation if obstacles will be solved 2 Supports cooperation but disagrees on some issues 3 Strongly supports cooperation

Power

1 Has a limited influence but his opinion is taken into account 2 Has influence on some issues

(33)

block the coalition

Salience

1 Issue of minor importance 2 Important issue but not a priority 3 Issue of high priority for the state

Table 1. Allocation of points

From this reasoning follows that the lower the difference in issue position between multiple players, the lower the conflict of interests and the higher the chance that cooperation will take place. Conversely, the higher the differences in issue position between multiple players, the higher the conflict of interests and the lower the chance that cooperation will occur. These effects are amplified by the power position of the states and the salience of the subject to the states in question. Under the right

circumstances, differences in power position and salience can make cooperation possible in a situation where states have a conflict of interests on a certain matter.

1.6.3. Power Asymmetry

States that engage in cooperation always have differing country characteristics, such as population size, military power, national income, territory, geographical position and access to natural resources. All these factors determine a states’ power to influence the course of water cooperation. This means that asymmetry in power is always present in cooperation. This can have both, a positive and a negative influence on cooperation. Asymmetry in power has a positive effect, when a stronger actor is interested in cooperation and uses his power to set up favourable conditions for cooperation. H. Milner argues that even “if a hegemon uses coercion to achieve benefits for itself, asymmetry in power still has a positive effect, because it

contributes to cooperative solution.”139 This is the case if one of the states is much more powerful than the others.140 But if the difference in power is less significant, its impact on cooperation can be negative.

The state with more power can have more means to develop projects, such as irrigation systems, hydropower stations, and uses more water, which limits the amount of water available to other countries.141 Cooperation in such a situation is

Water Diplomacy in Central Asia: Explaining the Lack of Cooperation