Great game visions and the reality of cooperation around post-Soviet transnational infrastructure projects in the Caspian Sea region

ABSTRACT

This article argues that the complexities of the Caspian Sea region and of the Baku–Tbilisi–Ceyhan (BTC) pipeline are only narrowly understood and explained by the geopolitical literature. In an attempt to address these shortcomings and to propose an alternative reading of transnational infrastructures, the BTC pipeline is revisited. Theoretically, the article uses insights from functionalism to explain the three phases of the BTC project: planning of the pipeline; construction of the pipeline; and, use of the pipeline. The core argument made is that the BTC project is much more than a power resource; it is the core medium of interaction for international and regional actors (governmental, non-governmental and intergovernmental) because of the number of connections that it makes possible in a highly contingent world. In contrast to Great Game assumptions, it is argued that transnational infrastructures are not necessarily negative or positive; rather, what is important is to view their contribution, mixed impacts and limitations from a broader perspective. In this sense, the article aims to situate the BTC infrastructure within the wider political, economic and social analyses of the Caspian Sea region.

KEYWORDS:

• Caspian Sea
• infrastructure
• multiple actors
• cooperation
• geopolitics
• BTC

1. Introduction

While the Caspian Sea (2004) region exhibits complex patterns of cooperation and conflict at all levels, ranging from the local to the global, it is often depicted as a prime example of a great powers’ arena due to its geographical location, rich natural resources and marked regional disputes (e.g. Alam 2002; Iseri 2009; Kim and Eom 2008; Li-Chen 1999; Shirin 2004). By using mostly realist doctrines (power/security/state), the vast body of research that exists on the Caspian Sea has framed highly complex infrastructure projects there through only a one-sided and exclusively geopolitical lens. In so doing, the literature has constantly used the metaphor of a “new Great Game” or “great power rivalry” to describe competitive energy relations and how great powers – states such as the United States, Russia, Iran, China and Turkey – may affect power constellations and strategic decisions within the region (Jaffe and Manning 1998; Karasac 2002; Kober 2000; Menon 1998).

In addressing Caspian Sea infrastructure developments, the literature focusses mainly on the role of states (external and local powers) and portrays every issue (environmental, technical, economic and social) from a purely geopolitical perspective (e.g. Iseri 2009; Kim and Eom 2008; Kleveman 2003). Within this discussion, Russia, Iran and Armenia have been identified as the main players behind every single issue or threat in the planning and construction phases of the Baku–Tbilisi–Ceyhan (BTC) pipeline (Bayulgen 2009; Ismailzade 2006; Talbott 1997). In the same vein, the project was expected to trigger rivalry and even war in the region following its construction (e.g. in Nagorno-Karabakh) (Bayramov 2016; Cohen 2002; Kim and Eom 2008). This understanding of the oil pipeline offers, however, the misleading impression that if there are natural resources, there will always be conflict and rivalry (Bayramov 2018; Klare 2001).

Against this established backdrop, the article posits that it is not sufficient to think about and scrutinise the BTC pipeline only in terms of rivalry and inter-state war. Viewing the Caspian Sea as a geopolitical battleground obscures the important layers of a more complex reality, as well as the underlying dynamics of material and non-material interdependency. Drawing on insights from classical functionalism, the three defining phases of the BTC project are analysed: planning of the pipeline; construction of the pipeline; and, use of the pipeline. The article thus pursues three core objectives: It first illustrates how during the planning phase there was great uncertainty with regard to the amount of natural resources extractable and concerning the foreign policies of regional states, which led to one-sided assumptions and exaggerations being made. Second, it explains how challenges such as financial, technical and social issues – beyond geopolitics – blocked the project and led to significant delays. In so doing, the article contends that without this network of multiple actors it would have been impossible to overcome these obstacles. Third and finally, the article explains the neglected role of the pipeline after its construction. More concretely, the article shows that – against Great Game expectations – the BTC pipeline has in a number of ways triggered functional and pragmatic cooperation among the Caspian Sea littoral states.

I select the BTC pipeline as the case study here for three reasons. First, considering its enormous economic, political, social and technical complexities, it represents one of the largest forms of energy infrastructure ever seen – requiring some 28,000 workers and consuming a substantial amount of the world's steel as part of its construction (British Petroleum 2016). Second, it is the first transnational project in the region since the end of the Cold War, involving several different countries (Azerbaijan, Georgia, Turkey). Third, it became operational in 2006, which makes it easier to analyse the long-term political and economic outcomes of the infrastructure project as well as to observe new developments occurring in the Caspian Sea region. The BTC pipeline crosses Azerbaijan, Georgia and Turkey to transport Caspian oil to Western markets. In this regard, the existing literature, has only discussed its long-term economic and political effects on these three key countries (e.g. Baran 2005; Frappi and Valigi 2015; Papava 2005). However, it is necessary to look beyond just these three countries and to include in the analysis other Caspian Sea states (e.g. Turkmenistan, Russia, and Kazakhstan) too as they also transport their oil through the BTC.

The article is divided into five parts. Following the introduction, the second section selectively introduces insights from classical functionalism. The third section explains the chosen methodology. The fourth and main section of the article discusses the identified three phases of the BTC pipeline. Fifth and finally, the conclusion presents the article's findings and identifies also its limitations.

2. Insights from functionalism

In order to establish the groundwork for the empirical part of the article, it is necessary to first selectively introduce the classical functionalist assumptions of David Mitrany (1966). The reason for this selectiveness is that there are already a number of works in existence that comprehensively address most principles of functionalism (e.g. Ashworth and Long 1999; Groom and Taylor 1975; Imber 2002; Rosamond 2005; Sewell 1966). Therefore, it is unnecessary to walk the same pathways once again. The selected points include, first, that states are neither unified actors nor the only ones operating in the international arena. Second, similar of problems challenge all sub-systems of the world alike, and individual states or blocs of states produce common interests which in turn act as incentives for seeking common solutions. Third, cooperation should be flexible, self-determined and pragmatic. Finally, transnational infrastructures should be viewed as mediating interfaces, creating new burdens, greater interdependency and facilitating interaction capacities.

I choose classical functionalism as the theoretical approach here specifically because the latest and updated liberal line of International Relations theories (e.g. regime theory, institutionalism and multi-level governance) fail to cover a number of key aspects necessary to explain the complex developments in the Caspian Sea. These fundamental aspects are, however, addressed by classical functionalism, namely the technical emphasise on material cooperation and material networks, as well as on taking non-state centric approach. The relevant liberal literature views cooperation and integration mostly as a set of intergovernmental treaties, the formation of institutions, formal and informal institutional agreements, different policy levels as well as organisational frameworks. Neglecting the material dimensions of cooperation, these theories fail to situate technical cooperation, material networks, and technology in broader political, social and economic analyses. However, taking into consideration the material and technological aspects, this article integrates contributions from scholars who do not see themselves as working in the functionalist tradition (e.g. Barry 2013a, 2013b; Coole 2013; Edward 2016; Latour 1996).

Only Petersen (2016) so far has used functionalism to explain the BTC project. Nevertheless, from an empirical perspective, Petersen (2016) covers only Azerbaijan, Georgia and Turkey, which overlooks the relationship between the littoral states. From a theoretical perspective, Petersen falls into a common trap and judges the functional developments in the Caspian Sea explicitly on the basis of a European benchmark. In this sense, he expects European-style cooperation and integration without recognising the Caspian Sea's historical, political, economic, material and normative differences. This article does not follow that European benchmark in discussing the Caspian Sea-style of cooperation, and hence recognises instead the distinct set of political, economic and social goals that are in play here.

First, compared to geopolitics, functionalism's distinctiveness lies in its ability to able to account for a more diverse array of actors (governmental, non-governmental and intergovernmental) (Rosamond 2005). It offers both specific and broad theoretical implications to cover the preferences, role, network and power of actors above and beyond just nation states. Multiple actors offer the required resources that most states lack, such as professional personnel, advanced technology, global networking, access to the world market, security guarantees and financial power (Forsgren 2008). In contrast to great game expectations, functionalists assume rather the prevalence of pluralist politics with multiple and diverse actors, who can and indeed do make contact and build coalitions across national frontiers and national bureaucracies. In short, they do many of the things traditionally – and exclusively – associated with the state (Mitrany 1966). Such a perspective is, in particular, conducive for understanding the involvement of multiple actors as well as their modes of cooperation vis-à-vis common technical, political and economic issues. The Caspian Sea region and the BTC project are subject to more complex and interconnected dynamics than pure high politics. To understand the complex enmeshment of the Caspian Sea and the BTC, and to illuminate the full picture here, it is necessary to understand also low politics. This theoretical perspective, therefore, is helpful to understand who is behind these dynamics and how they talk about these developments.

Second, in contrast to geopolitics, functionalists have pointed out that not all games between actors are zero-sum ones. Interaction is often better characterised as positive-sum games, in fact. This approach is useful because it does not completely dismiss the central aspect of rivalry between different parties but rather seeks to draw a circle around them. The main reason is that functionalism aims to first establish mutual trust between conflicting groups or search for possible alternatives. In this regard, functionalism argues that cooperation between multiple actors should be flexible and pragmatic (Mitrany 1966). It is important to determine those activities that are common, where they are common and the extent to which they are common. Also, actors are part of these activities with respect to their interests and resources – and there should not be an obligation to participate in all of the activities or contrariwise to stay out of them (Mitrany 1966). This theoretical perspective is, in particular, helpful for understanding the changing conditions and preferences of actors in the Caspian Sea region.

Third, functionalism situates technical cooperation in broader political, social and economic analyses (Rosamond 2005). The notion of a specific “technical sector” can be interpreted as “infrastructure” such as railways, telecommunications, roads, cables and pipelines, which transcend national boundaries, connect a number of different actors, bring new restrictions and affect interaction capacities. A technical sector is also self-determined, which means it fixes its geographic scope, actors, organisational structure, inherent nature, boundaries and power. For the objective of this article, this is a promising insight since it considers complexity and the role of transnational infrastructure projects. Both geopolitics and similar liberal lines of IR theorisation fail to include this fundamental aspect, and thus to situate technical cooperation, material networks and technology in those aforementioned broader political, social and economic analyses.

Unlike geopolitics, functionalism explains how technical issues encourage multiple players to cooperate as these issues are beyond the political, economic and physical capacities of any single actor to address (Mitrany 1966). More specifically, to build a transnational infrastructure requires important economic, political and technical investments. Due to these costs and the matter of income generation, functionalism argues that cross-national technical resources increase the burden of conflict in the region – which would make international disputes less beneficial, less popular and ultimately less feasible. States would think twice when thinking about starting a war with their neighbours (Oneal et al. 1996). However it is important to highlight that cross-border infrastructures per se do not necessarily prevent conflict; the condition of interdependency also has to be recognised and understood by the concerned actors. If they are not aware of their interdependency, then conflict may still arise even when it is not in their best interest.

Finally, contrary to geopolitics, functionalism argues that technical cooperation enhances the chances of material and non-material interdependency emerging between different actors. From a non-material perspective, transnational technical resources require relaxed sovereignty principles, common rules, established management strategies and firm regulatory frameworks (Henrich-Franke 2014). To realise these, actors need to have regular meetings, joint studies and dialogue between politicians, experts, the media and citizens, which will help them to learn about each other's different perspectives, share their risks and deal with them accordingly. This might also increase socialisation among different actors, and may lead to the creation of joint institutions to govern transnational infrastructures. From a material perspective, meanwhile, technical resources are mediating interfaces, as they allow people, things and signs to travel across space by means of more or less standardised pathways and protocols for conversion or translation. For example, Mitrany favoured the Universal Postal Union (UPU) because, as one of the world's global service organisations, the UPU brought different actors together under common rules, regulations and agreements for specific material exchanges –letters, mail and parcels. For the purpose of this article, this theoretical perspective is particularly beneficial for explaining how the BTC has led to the establishment of material and non-material interdependencies among diverse actors in the region.

However, there is still room for the enrichment of functionalism in certain areas. One important enhancement would be to specify for whom something is functional, and to what end. In other words, functionalism should be more actor and value specific (De Wilde 1991). The literature has equated functionalism with “in service of humanity” or “instrumental to the solution of a social problem” and not “in service of a ruling elite” or “instrumental to a specific organisation and its bureaucracy”. Despite these shortcomings, the functionalist insights offer helpful tools for capturing the complexity of the BTC infrastructure and the underlying logics of Caspian Sea cooperation. By using these functionalist tools, this article will able to reveal the actors behind the BTC pipeline, their roles and preferences, any common issues as well as the material and non-material functions of this pipeline project in the Caspian Sea.

3. Methodology

This study is based on primary sources, interviews, field research, scholarly journals, policy documents, international reports, public debates resonating in the media as well as political speeches and newspaper articles. Even though the article covers the time frame of 1994 to 2012, it is not a purely historical account of events. Twenty-two semi-structured and in-depth interviews were conducted with local and international experts holding different positions within 14 regional and international institutions, individuals both implicitly and explicitly involved in the BTC pipeline. The interviews were conducted in two languages, English and Azerbaijani. The semi-structured interview model was selected because it is formal enough to ask the same set of questions, but sufficiently flexible that ones spontaneously arising could also be worked into the conversational flow (Bryman 2016; Hennink, Hutter, and Bailey 2011; Mosley 2013). By doing this, I could ask new questions that followed up on interviewees’ replies, and could vary the order of my queries. In-depth interviews allow for listening to the experiences, interpretations and feelings of an expert (Bryman 2016; Hennink, Hutter, and Bailey 2011).

I undertook two field research trips in Azerbaijan, Baku. The first from November until December 2017, and the second in May 2018. Due to the limitations of time and financial resources, I conducted several Skype interviews with experts from Armenia, Georgia, Iran, Turkmenistan and Turkey. Interviews lasted between 30 and 90 minutes each. During the interviews, I strived to type up the main points but I might have missed certain peripheral ones. It is worth mentioning that for several reasons the data gathered from these interviews is presented here anonymously (they were not recorded either). The first is confidentiality, which was mutually agreed upon at the beginning of each interview. Second, this approach protects respondents from retaliation for divulging potentially controversial information. Third, it can encourage openness, as people often speak their minds if they no longer have to worry about their statements coming back to haunt them. Fourth, individuals were not speaking on behalf of their institutions and were instead giving their personal opinion, making institutional affiliation less relevant. After all interviews were transcribed and I developed a coding scheme using the qualitative data analysis software, Atlas.ti.

In terms of challenges and limitations, there were certain hurdles to be faced during the field trips themselves as well. First, certain interviews reflected only the official policy line and therefore were not entirely useful or innovative. I was able to obtain innovative data from a few interviews, nevertheless. Second, it was difficult to gain access to certain government institutions – such as the Ministry of Foreign Affairs and the Ministry of Defence – and certain questions required access to the top level of decision-making; therefore, most interviews took place in a private or informal setting. In this sense, I could not access experts from two Caspian states: namely, Kazakhstan and Russia.

4. Discussion: planning, construction and post-construction stages

4.1. Planning of the BTC pipeline

On 20 September 1994, after three and a half years of extensive negotiations, Azerbaijan and a consortium of foreign oil companies signed a Production Sharing Contract (PSC) in order to develop Azerbaijan's Azeri–Chirag–Gunashli (ACG) deepwater oil reserves – this would later come to be known in the media as the “contract of the century”. Following this, a number of pipeline routes were initially explored to transport the oil to international energy markets – including one going east from the Caspian Sea to China, another heading south to Iran and a further one extending the existing pipeline connections of Baku–Novorossiysk.¹ However, the economic sanctions imposed by the US on Iran, the poor state of the existing Baku–Novorossiysk pipeline, and the fact that other routes tended to terminate at the Black Sea and required oil to transfer through the Bosporus (already congested with tanker traffic) together all resulted in preference for a route through Baku–Tbilisi–Ceyhan (Figure 1).

Figure 1. Current pipelines from Azerbaijan. Source: The State Oil Company of Azerbaijan Republic (SOCAR) (2018).

Note: Figure 1 includes three pipelines: the BTC, the Baku-Tbilisi-Erzurum and the Baku-Novorossiysk pipeline.

The agreement was signed in Istanbul on 18 November 1999 by the presidents of Azerbaijan, Turkey, Georgia and Kazakhstan in the presence also of the US president. This pipeline links Sangachal Terminal, situated on the shores of the Caspian Sea, to the Ceyhan marine terminal, lying on Turkey's Mediterranean coast. The pipeline first became operational in June 2006, being run by BP (Table 1).

Table 1. Timeline of the BTC Pipeline from 1994 until 2006.

Source: Sovacool and Cooper (2013, 111).

4.1.1. Geopolitical uncertainty

From the very first day of the BTC project's existence, the Caspian Sea would be described as a new geographical site of the Great Game (Jaffe and Manning 1998; Karasac 2002; Kober 2000). By using nineteenth-century geopolitical thinking, the literature describes Russia and Iran as the main rivals to the development of any pipeline here. On the other hand, the European Union and the US are described as saviours. Normally, the Russians should not have taken any notice of relatively marginal amounts of oil being produced by Azerbaijan. Russia's overall production of the commodity is significantly greater in volume – generating some 500–540 million tons of oil (Henderson and Grushevenko 2017). Current BTC production does not even amount to 5 per cent of this output level. For example, according to BP, from 2006 until 2018 the BTC pipeline carried just three billion barrels of oil to the Ceyhan marine terminal in Turkey. Considering this, one may argue that neither Russia nor the other great powers should worry about the amounts of oil produced in Azerbaijan. Why, then, was the planning phase still mainly dominated by geopolitical assumptions and exaggerations?

First, in the early 1990s as well as first years of the new century a number of events played a key role in creating the grounds for geopolitical uncertainty. For example, because of the Nagorno-Karabakh conflict between Azerbaijan and Armenia, the pipeline route was not sufficiently secure. Furthermore the legal status of the Caspian Sea was not clear, and as such the littoral states were struggling to decide on it. Due to this uncertain legal status, it was also difficult to determine the ownership of several oil fields at sea – namely, Araz, Alov and Sharg. Consequently, there was ongoing disagreement between Azerbaijan, Iran and Turkmenistan as claimants.

Second, when reviewing the literature, newspapers and official speeches published throughout the 1990s, one realise that large amounts of scholarship were devoted to predicting the Caspian Sea's proven energy reserves – and, moreover, tended to cite reserve figures that ranged from the optimistic to unrealistic (e.g. Alam 2002; Bahgat 2002; Jaffe and Manning 1998; Kim and Blank 2016; Ruseckas 2002). The most commonly used estimate for the region's oil reserves was 200 billion barrels, with no distinction made between “proven” and “possible” reserves in the late 1990s. For example, in July 1997 US Deputy Secretary of State Strobe Talbott described the Caspian Sea oil reserves as being “as much as two hundred billion barrels of oil” (Kleveman 2003, 7). Later, US secretary of state James Baker would go even further: “Caspian oil may eventually be as important to the industrialised world as Middle East oil is today” (New York Times 1997). In the late 1990s, a number of academic articles only strengthened these exaggerated assumptions. For example, Pipes argued that the Caspian region holds “oil reserves estimated to be at least as large as those of Iraq and perhaps equal to those of Saudi Arabia” (1997, 73). These exaggerations may seem unimportant. But in some cases, particularly when the overly optimistic figure of 200 billion barrels is wrongly compared to total global reserves of about one trillion, it can enhance international attention paid to the region in the short term and unwittingly cause subsequent conflicting understandings, deep suspicion about motives and information struggles (Conca 2001).

By combining these exaggerated reserves with the events of the early 1990s, the scholarship predicted unrealistic scenarios for the region. For instance, due to the uncertain legal status and to the presence of potentially vast oil fields, international experts argued that this dispute between the littoral states may eventually reach the point of military means being used should it continue further (Haghayeghi 2003; Lelyveld 2001). Once Azerbaijani oil began to be exported to world markets in a stable manner through the BTC pipeline, it was suggested that there existed a significant chance that Azerbaijan would begin a new war with Armenia (Bayulgen 2009; Kim and Eom 2008). Along the same lines, Cohen argued that “supporting Armenia, Iran or Russia could disrupt or sabotage the BTC pipeline” (2002, 5). Using outdated geopolitical assumptions, Dodds (2005) described the Caspian Sea basin as part of a gigantic strategic triangle (along with the South China Sea and West Asia) that would come to shape the patterns of potential (resource) wars in the twenty-first century. In the early years of the new millennium, Russian energy company Lukoil became one of the BTC pipeline's shareholders; it was thus claimed that it became part of this project due to Russian political pressure. But, Lukoil quit the project in 2002 for economic reasons. Naturally, it was again argued that Lukoil wanted to sabotage the project in selling its shares (Ismailzade 2006).

Considering this, it can be concluded that in the early 1990s there was a tendency to present every development through the lens of geopolitics and great power manoeuvring. In using inaccurate information, much of the analysis that has been conducted is of dubious standard – with facts being accepted without question, and used without being subject to any semblance of academic rigour. Therefore, a false and misleading image of the BTC pipeline in the modern Caspian Sea has been created. Such outdated nineteenth-century geopolitical thinking is unavoidable, but awareness about it needs to be raised as it leads to new insights regarding the shortcomings of dominant academic and political practices. These examples show that according to the relevant literature, if there are natural resources there will also be conflict and rivalry. But this is only one possible option; another is the opposite outcome, because cooperation might, in fact, also emerge due to natural resources.

4.2. Construction of the BTC pipeline

4.2.1. Challenges besides geopolitics

Throughout the construction phase, the pipeline faced a number of technical, environmental and economic barriers, namely engineering failures, unstable oil prices and social protests. The following section explains how these challenges influenced the development of the project during that construction phase.

4.2.2. Technical challenges

The first technical challenge faced was unexpected engineering failures. In 2004 a number of consultant engineer companies reported problems with cracking in the pipeline due to the coatings used in the Azerbaijani and Georgian sections of it (The Guardian, 17 November 2004). WorleyParsons, the lead independent consultant engineering firm involved in the project, criticised the inaction of the BTC project management team, which had allowed the problems to become greater than necessary. This issue increased the concerns of international investors backing the pipeline. Among them, Banca Intesa (Italian Bank) had expressed apprehension about the technical problems and later dropped the BTC project completely for this very reason (Financial Times, 1 December 2004).

Similarly, the dispute over the significance of SP-2888 coating² became particularly intense during the course of an enquiry by the United Kingdom Parliament's House of Commons Select Committee on Trade and Industry into the activities of the UK government's Export Credit Guarantee Department (ECGD). It was found that SP-2888 is not a high-quality material, due to the technical issue of cracking and to a strong campaign by human rights and environmental non-governmental organisations (NGOs). The UK Parliament devoted considerable time to the failed use of this material, and it also questioned the involvement of BP and the ECGD in this project (Barry 2013, 143). SP-2888 was the primary issue that was discussed in the House of Commons regarding this. After all, problems with the coating material could not be denied, and BP accepted that they had happened. Long sections of the pipeline had to be repaired as a consequence of the failure of the coating material, which increased the overall cost of the project.

Additionally, harsh weather conditions and unexpected archaeological findings created further technical challenges. The BTC pipeline corridor climbs gradients as high as 3,000 metres steep in some places, and is almost 200,000 kilometres long – with increased pressure being required to move oil up and down inclinations and slopes (Pipeline and Gas Journal 2006). It was argued that unexpected snowstorms and harsh weather conditions made some parts of the pipeline inaccessible for up to four months, which accounted for delays at an extra cost of USD 270 million (Pipeline and Gas Journal 2006). Finally, once construction began contractors encountered more archaeological sites and unexplored places along the pipeline route than the planners had anticipated – which also led to delays and extra technical work and financial costs.

Considering these technical challenges, it can be argued that it was not Russia, Iran or Armenia but the issues themselves that led to significant delays during the construction process – and meanwhile increased the overall cost of the BTC project. These underestimated technical issues forced the UK Parliament to question its economic and political support for the pipeline. In this regard, the project would never be completed if one of the powerful members of the EU withdrew its support for it.

4.2.3. Economic barriers

The pipeline project additionally faced a multitude of economic hurdles. During the planning stage, project sponsors assured the governments of Azerbaijan, Georgia and Turkey that the BTC pipeline would cost approx. USD 2.1 billion; then it was increased to USD 3.6 billion, while in the end it came to cost approximately USD 4 billion to build (BP Report 2012). Sovacool and Cooper (2013) argue that much of the cost overruns resulted from underestimating the expense of environmental and social impact assessments. In the year 2000 three members of the consortium – Lukoil, ExxonMobil and Penzoil – withdrew from the project for economic reasons. As mentioned earlier, Italian investor Banca Intesa also pulled out of the consortium in 2004, which led to construction delays and extra financial outlays. Following this event, the head of SOCAR said delays to construction work on the project could increase its costs by about USD 400 million. It is important to note that every day of delay cost BP potential oil revenues (Pipeline and Gas Journal 2006). Additionally, in 2004 the price of oil was less than USD 40 per barrel, creating pressure to complete the project before the commodity's value bottomed out entirely (OECD Data 2018).

Another economic challenge was the land acquisition and compensation process for the BTC pipeline. To address this process and its impacts, the BTC project needed a comprehensive and well-structured programme. However, according to an International Finance Cooperation (IFC) report: “BTC contractors underestimated the scale and complexity of the land acquisition process and how much lead time and resources this would require in countries where land registration systems and land records were weak or non-existent” (2006, 19). Hence in many cases the BTC project had to start from scratch, initiating land survey work and identifying thousands of rights holders – which ultimately took considerably longer than originally anticipated and therefore induced additional spending requirements and construction delays. Because of these economic issues, the BTC pipeline ended up costing, as noted, USD 4 billion. Considering Azerbaijan's and Georgia's weak economy at the start of the new century, it can be argued that these extra economic costs created an acutely difficult situation for them. Other international partners questioned the profitability of the project due to these issues. To attract new investors to it was also not easy as Azerbaijan and Georgia were considered too risky due to their own regional conflicts and internal issues.

4.2.4. Social and environmental challenges

Throughout the construction phase, active protests by a number of NGOs and environmental grassroots movements were another core challenge that the BTC project stakeholders had to confront. Organisations such as Friends of the Earth, Kurdish Human Rights Project and CEE Bankwatch were staunchly opposed to the pipeline project. They sent letters to high-profile members of the World Bank and BP, and organised protests at the offices of the European Bank for Construction and Development (EBRD) (Carroll 2009). In August 2002, a coalition of NGOs released a series of fact-finding reports based on investigative missions to Turkey, Azerbaijan and Georgia, excoriating the project for the threat that the pipeline's construction and operation posed to both human rights and the environment (e.g. risks of oil spill and loss of biodiversity) (Amnesty International 2003; Carroll 2012; Molchanov 2011). In 2004, Tamar Libanidze, Georgian environment minister, halted the construction of the Borzhomi section in that country due to these environmental risks (Burton 2004).

Another barrier was social protests. The BTC venture was hit by worker discontent in 2004: labourers in both Azerbaijan and Georgia argued that the companies involved in the laying of the pipeline were engaging in unfair workplace practices. On 28 February 2004, about 400 workers employed by the Greek-based Consolidated Contractors International Co. went on strike in the Kurdamir District of Azerbaijan, due to perceived social injustice and discrimination both in terms of wages and ethnicity (Appelbaum 2004). These protests played a significant role in slowing down the project, increasing political attention and raising expenditure on it.

Above all, these examples illustrate the central role of technical, economic and environmental issues – which increased the costs of the project, led to delays, gave rise to investigations and, indeed, almost put an end to it entirely. Considering these challenges, one may argue that it was not Russia, Armenia or Iran but one of the BTC pipeline's key stakeholders, Georgia, who blocked the pipeline's construction because of environmental issues. In the same vein, it was one of the core Western supporters of the project, the UK, who investigated and questioned the involvement of BP and ECGD in the project for their technical and material failures. Finally, the project was delayed by the mass protests of workers, NGO pressure and by land acquisition matters. The project was halted because people protested against their measly salaries and harsh work conditions; none of the great powers of the region pushed them to do this. These neglected material and non-material issues created obstacles for the BTC project, threatened its success and increased the overall cost from USD 2.1 billion to USD 4 billion. From a functionalist perspective, these examples also indicate that political agreement is not the only decisive condition for the realisation of complex infrastructure. Rather, non-political requirements also need to be fulfilled.

4.2.5. Network of multiple players

These combined technical and geopolitical challenges have produced an environment where Azerbaijan, Georgia and Turkey must now rely on the involvement and coordination of multiple actors, namely international NGOs (INGOs), NGOs, financial institutions and transnational corporations (TNCs). As explained in the theory section, the presence of two or more actors in transnational infrastructure can reduce serious financial, political and security risks – which such projects often encounter. This section illustrates, therefore, exactly how a network of actors with varying interests created systematic and functional coordination in the context of the BTC project.

4.2.6. Oil companies

One of the key actors herein are multinational oil companies, such as BP, Chevron, SOCAR, Inpex and Statoil Hydro, who between them offered a number of the required resources to transport landlocked oil to international markets. Due to their economic power, oil companies’ investments are a strategic source of revenue and a key input to the budgets of the countries of the region. For instance, because of economic troubles related to the collapse of the Soviet Union, Azerbaijan's and Georgia's economic growth contracted by almost 60 per cent between 1990 and 1995 (Sovacool and Cooper 2013). The Azeri and Georgian governments were therefore very receptive to earning substantial transit revenues from a possible pipeline, and they were willing to do their best to attract the participation therein of international oil companies (Kalyuzhnova 2008). According to one interviewee, “in the 1990s, Azerbaijan was willing to accept a ceasefire with Armenia to create a safe investment environment for international oil companies. In light of this, Azerbaijan was able to attract crucial energy companies to the pipeline project”.

The BTC is owned and operated by a consortium of 11 international oil companies, being managed overall by BP (BP 2017). In the early 1990s, forming such a consortium was not easy for companies since they were also in competition with each other during the tendering and negotiation phase of the BTC pipeline. One of the key issues was whether and if so how to include Russian energy companies. Nevertheless, Western energy companies were aware that the financing, resources and political realities in the Caspian Sea region required working together. Therefore, they sought to accommodate Russia by giving Lukoil a 10 per cent stake – as the success of their project depended on good relations with Russia (Edwards 2003). Overall, the involvement of these companies contributed to economic growth in Azerbaijan, Georgia and Turkey by providing a substantial new revenue stream. They helped the regional countries to establish unprecedented commercial links with the rest of the world.

Besides their economic leverage, the involvement of Western energy giants comes with political and security advantages too. They put Azerbaijan and Georgia on the map in terms of attracting foreign direct investment and gaining Western support for their sovereignty, resolving territorial conflicts and for ensuring security. For example, according to a local expert in Baku, “due to the involvement of international oil companies, some Western countries (for example, the UK and the US) have paid more attention to the region's conflicts”. This attention was very important for Azerbaijan and Georgia, as it could help them to keep up diplomatic negotiations and prevent further violent clashes. Additionally, one interviewee mentioned that the “active involvement of TNCs offers extra security to the BTC because it is the property of both Azerbaijan and the West”.

In addition, BP has implemented significant security measures along the energy route, mainly in the form of patrolling and monitoring. An expert from the company said that although protection of the pipeline is ultimately the responsibility of the relevant governments, BP is involved in addressing this security matter with its own measures too – such as providing advance technology, training guards, implementing social projects, and offering financial support. In Azerbaijan, BP has implemented facility protection and security guard services through its private security provider, Titan D, while closely cooperating with the Export Pipeline Protection Department – the Azeri government agency appointed for infrastructure security. Besides these measures, BP has also launched several social programmes (e.g. repairing roads, supporting agriculture, educational initiatives etc.) along the pipeline's route to support local villages and gain their support. According to a representative of the company, in this way BP can cooperate with local people and they inform government officials in advance about any terrorist or sabotage plans. Table 2 below outlines the states, companies and private lenders involved in the BTC project.

Table 2. Actors.

Source: Authors own compilation.

Finally, the exploitation and transportation of oil from the Caspian Sea would not be possible without modern technology, which the regional states lack. In this sense, the consortium companies revitalised the technical capacities of Azerbaijan by offering advanced oil and gas processing plants and fabrication facilities. They contributed to the upgrading of local experts’ knowledge by offering a number of educational and capacity-building training programmes. In line with the theoretical proposition of this article, BP's contribution to the BTC project illustrates, therefore, that it offers many of the things that regional states lack and were not otherwise able to attain. It is, however, important to note that by involving themselves in the project, these companies have also benefitted significantly from it – by adding new reserves to their resource bases, by exploiting vast natural resources for significant profit and by diversifying their portfolios away from reliance on fields in Alaska, the North Sea and South America.

4.2.7. Coordination with the World Bank Group: EBRD, IFC and private banks

Despite the heavy investment of BP and other energy companies, covering all of the costs for this massive project has still required funding also by international banks and financial INGOs – such as the World Bank, the IFC, the ECGD as well as the EBRD. Ensuring sound coordination between them was decisive to securing sustainable funding and reducing attendant political risks. Nevertheless, in the years from the end of the Cold War into the new century the regional countries lacked crucial lobbying and networking experience. Furthermore, ones like Azerbaijan and Georgia were – as noted earlier – considered too risky for international banks and financial institutions to invest in. As such, one interviewee noted that:

By using their access to global donor networks, the consortium companies – particularly BP – facilitated relations between Azerbaijan and financial institutions: the World Bank, EBRD, ECGD, EXIM Bank and IFC. BP has played a key role in all stages of the BTC project since the 1990s. It is one of the strong and popular European energy companies, and its involvement attracts other Western financial institutions and gives them more security and reliability.

These global financial institutions invested in the BTC pipeline as a way of helping Azerbaijan, Georgia and Turkey graduate to the global economy (Petersen 2016; Sovacool 2010; Sovacool and Cooper 2013). For example, EXIM Bank was only one of seven countries’ export credit agencies involved in financing the project. From 2003 to 2005, EXIM Bank had to approve financing of up to USD 160 million to help complete the project (Bashir 2017). The IFC provided an overall investment expenditure of USD 250 million for the development of the BTC pipeline. Additionally, in 2003 the EBRD approved a 12-year loan of up to USD 125 million for the BTC project itself and syndicated a 10-year USD 125 million loan to commercial lenders (Pyrkalo 2016). In December 2003, the ECGD approved a line of credit for the project of USD 450 million (GBP 81,703,893). According to Barry (2013), the involvement of the ECGD in the project was intended to reduce the financial risk to investors – but also helped to ensure that the UK government in particular would have a direct interest in the eventual completion of the project.

Additionally, governments that received loans from the World Bank and other financial institutions are obliged to implement a package of reforms (relating to environmental, technical and economic standards). Considering the pressure coming from different NGOs, these organisations worked with the BTC consortium companies and three governments to help with land resettlements, fostering local businesses and ensuring environmental compliance. To optimise Azerbaijan's management of its resource wealth, the World Bank Group advised the Azerbaijani government on the creation of the State Oil Fund of Azerbaijan (SOFAZ). To promote transparency, the Group facilitated the country's participation in the Extractive Industry Transparency Initiative (EITI); Azerbaijan would, however, leave this institution in 2017, due to its human rights demands. International oil companies and financial institutions together helped Azerbaijani and Georgian energy companies to transform from Soviet-era entities into more transparent, modern state-owned enterprises (Petersen 2016).

After becoming involved in the BTC, these actors offered their strong networks to help attract further financial support (private banks) and to mitigate any political risks. Nevertheless, and as pointed out earlier, it has been argued that “the private capital markets were not willing to loan to countries like Azerbaijan and Georgia because they are too risky” (Sovacool and Cooper 2013, 118). However, this situation changed thanks to the strong networks of the EBRD and IFC – who were able to guarantee the cheap lending of financial products to Azerbaijan and Georgia.

The two countries utilised these organisations’ vast network of financial, social and environmental experts to ensure the minimisation of costs and maximisation of assets. They played the much-underappreciated role of risk mitigators in the process of opening up the Caspian Sea's riches, assisting in attracting the private leaders of the financial world – including Citibank, ABN Amro and Societe Generale – to help finance the remaining (minor) outstanding amounts (Carroll 2012). In line with the theoretical proposition of this article, this example illustrates that the power of actors rests in large part on their access to global donor and technical assistance networks that regional actors may not be able to otherwise reach (Mitrany 1966). The export credit agencies and bilateral financial institutions of the US, the UK, Japan and others teamed up with the EBRD and IFC to lend more than just a hand. In total, the funding model proposed put up USD 1.7 billion of public money for the project in a 70/30 debt/equity structure (Carroll and Jarvis 2014). It has been claimed that 70 per cent of the project costs were funded by a group of lenders that included the World Bank Group's IFC, EBRD, the export credit agencies of seven countries as well as a syndicate of 15 commercial banks (IFC September 2006).

These empirical findings reinforce the functional proposition that technical issues are beyond the political, economic and physical capacities of any individual actor to solve – thus requiring the involvement of multiple players. In line with this, it can be argued that the BTC project would not have been completed in a timely manner if a number of different actors were not part of the project. They pooled their resources to deal with specific issues. The legitimacy and supervision provided by INGOs like the World Bank Group and by private companies were critical to securing public recognition and support from the US and the EU governments for the project (Petersen 2016). Although the regional countries are still witness to weak rule of law, endemic corruption and limited institutional capacities as well as transparency, these actors between them ensured that ultimately the BTC pipeline was successfully built according to certain Western standards.

4.3. Use of the pipeline: pragmatic cooperation and regional conflicts

When addressing the BTC project's socio-economic impact, the existing scholarship focusses mainly on the trilateral relationship between Azerbaijan, Georgia and Turkey or, alternatively, competition between the littoral states (e.g. Bayulgen 2009; Dikkaya and Ozyakisir 2008; Frappi and Valigi 2015; Mammadyarov 2007). However, despite the diverse literature on show, less research has been done explaining how the BTC pipeline has led to cooperation among the Caspian Sea states since its construction.

4.3.1. Pragmatic cooperation

Unlike the geopolitical assumptions, which predicted naval conflict and rivalry, the Caspian Sea countries actually started to show keen interest in the BTC route following the successful completion of the project. The first such example of cooperation is Turkmenistan, which from July 2010 onwards started to transfer its oil through the pipeline. Despite the lack of consensus between Turkmenistan and Azerbaijan over the Azeri–Cirag–Kepez oil fields – which the great powers literature argued may eventually lead to (naval) warfare between the two countries – the BTC pipeline has, in fact, provided these two countries with the opportunity to enhance their cooperation. According to BP Azerbaijan, the pipeline is capable of handling some 800,000 barrels per day and Turkmen oil accounts for 4–5 per cent of this flow volume (Radio Free Europe, 12 August 2010). For example, 371,206 tons of Turkmen oil were transported via the BTC pipeline in June 2016 alone (Ismailova 2016). Since there is no pipeline connection between Turkmenistan and Azerbaijan, oil is brought to Baku aboard tankers. By joining the BTC pipeline, Turkmenistan has diversified its oil export options to include one that does not pass through Russia and has also secured access to international energy markets through an alternative route. These days, Azerbaijan and Turkmenistan are furthermore in the process of discussing the transportation also of Turkmen gas to Europe through the “Southern Gas Corridor”.³

As a second example, on 16 June 2006 Kazakhstan officially joined the BTC project – for which an agreement was signed in Almaty by the presidents of Kazakhstan and Azerbaijan (Radio Free Europe, 16 June 2006). Due to the absence of an existing pipeline between the two countries, Kazakh crude oil is shipped to Baku across the Caspian Sea and then pumped through the 1,770-kilometre-long BTC one to Turkey's Mediterranean port of Ceyhan. Following this, in November 2008 the national energy companies of Azerbaijan and Kazakhstan concluded an agreement with respect to the development of a Trans-Caspian oil transport system to help get Kazakhstani oil to international markets (The Moscow Times 2008). The network would be initially able to ship 500,000 barrels of oil daily (23 million tons a year), eventually increasing to 750,000–1.2 million barrels per day (35–56 million tons annually). The new Trans-Caspian oil transport system agreement and the commodity's shipment via the BTC pipeline indicate that, despite some disagreements over transit tariffs and the use of Black Sea terminals, Azerbaijan and Kazakhstan are willing to cooperate on moving the latter's oil – and that, moreover, Kazakhstan is keen to improve its export capacities and is looking for options vis-à-vis diversifying export routes (Guliyev and Akhrarkhodjaeva 2009).

Kazakhstan has still managed to restore the pumping of hydrocarbon resources via the BTC pipeline, but not the Kazakhstan Caspian Transportation System (KCTS) project. The Tengizchevroil company provides for an annual exportation of four million tons of oil via Azerbaijan, of which three million should be transported specifically via the BTC pipeline (Parkhomchik 2016, 143). According to the Azerbaijani State Statistics Committee, over 6.53 million tons of transit oil (from Turkmenistan and Kazakhstan) were pumped through the BTC pipeline in 2018 (Caspian Barrel 2016, 27 April 2018).

Furthermore, a week before Kazakhstan officially committed to the pipeline an Iranian official said Tehran wanted to explore the BTC export option too. According to Mahmoud Khagani, the chief of the Caspian Sea Department of the Iranian Oil Ministry “we are currently exploring for oil in the southern Caspian Sea. Our relations with Azerbaijan have been developing so successfully that, if we get positive results in the southern Caspian, we could discuss possible cooperation” (Eurasianet, 18 July 2006). Although this aim was not fulfilled due to reasons of it being commercially unprofitable and of limited natural resources in the southern section of the Caspian Sea, this political statement illustrates that Iran was also keen to seek new options regarding reaching Western energy markets. In 2018 Tehran and Baku discussed the possibility of establishing a joint oil company for the exploration of natural resources in the Caspian Sea (Reuters 2015; Trend, 15 May 2018).

Finally, in 2009 Russia's largest oil producer, Rosneft, became interested in options to export oil through the BTC pipeline. Rosneft's president, Sergei Bogdanchikov, told the press: “If the project meets the economic interests of both sides, naturally we will be able to export our oil through the BTC” (Azernews, 23 September 2009). In response to this, Rovnag Abdullayev, president of SOCAR, said: “If an appeal is received, it may be considered, and even its realisation in the future is possible” (Azernews, 23 September 2009). However this idea has, to date, ultimately not moved beyond abstract statements, because it is argued that the BTC pipeline is economically less appealing for Russia (Daly 2014).

Despite this, negotiations have continued between SOCAR and Rosneft from time to time (Antidze 2014). Moreover, in 2014 Lukoil showed interest in transporting its oil via the BTC pipeline. Later that company announced, on 16 May 2014, that its oil would soon be delivered to Europe via the BTC pipeline (Caspian Barrel, 18 May 2014). In the same month Lukoil delivered a trial batch of 30,000 tons of oil via the pipeline. For economic reasons, it started to transport Russian crude oil and condensate (Lukoil) in 2018 (British Petroleum 2018; BP First Quarter 2018 Results). Since it constitutes a commercial secret, uncovering the exact amount of transported Russian oil is not possible. During my fieldwork interviews, a local expert from Azerbaijan posited that Lukoil intends to transport approximately 500,000 tons of oil this way.

These empirical findings reinforces the theoretical proposition that post-construction the transnational BTC infrastructure has changed the dynamics of the Great Game in the region, by offering a functional system for pragmatic cooperation. It is indeed true that infrastructure becomes “strategic” because of the number of connections that it makes possible in a highly contingent world (Latour 2005). In this regard, the completion of the BTC pipeline has changed the preferences of the Caspian Sea states, enhanced regional interaction capacities and connected landlocked countries to both global and regional networks. Existing uncertainty and hostility have decreased to some extent, which has furthermore enhanced the interest of the littoral states in the project. Cooperation between them is occurring because all three stand to benefit from it, as each possesses a resource that the other two lack of their own. Both energy and transport are valuable enterprises that promise to bring financial reward. It is a pragmatic, flexible and technocratic cooperation unfolding, as the littoral states can be part of these activities with respect to their interests and resources and there is no an obligation to either participate in the project or to stay out of it.

Having said all this, it is worth noting that the downside of infrastructural cooperation is that it is set up to exploit specific natural resources or geographic places. In this case, Baku is one of the few sites to have been exploited in this context. Due to this fact, it is the Caspian Sea capital cities that constitute the main beneficiaries of ensuing economic development, while other ones are ignored and miss out. Therefore, such cooperation could also eventually lead to the exploitation of the periphery rather than to its development.

5. Conclusion

This article has challenged the Great Game visions that depict transnational infrastructure as one of the core sources of geopolitical rivalry and conflict in the Caspian Sea region. Drawing instead on insights from classical functionalism, it has been argued that the BTC pipeline should not be viewed in isolation; rather its complexity and sophistication need to be fully taken into consideration. This is because, first and foremost, the BTC project is not just made up of metal pipes and oil, but also of the extensive legislation, logistics and state and non-state authorities that also all help support it. The BTC pipeline has brought multiple actors together, connected national borders, increased the cost burdens of regional conflict and altered the structure and landscape of the Caspian Sea region.

By using tools from functionalism, the article has explained the three phases of the BTC pipeline project: planning of the pipeline; construction of the pipeline; and use of the pipeline. The article, first, showed that the planning phase was dominated by geopolitical assumptions due to the exaggeration of the extent of natural resource reserves as well as to political, economic and technical uncertainties. This enhanced international attention given to the region in the short term, and also unwittingly caused subsequent conflicting understandings of it, deep suspicion about motives and information struggles. Thus, a false image of the region has been created.

Second, it has been illustrated here that geopolitical arguments do not reveal the full picture when it comes the BTC project. By discussing the construction phase, the article has shown that analysing related challenges (e.g. environmental, technical economic and social) is an inadequate approach when it comes to Russia for example. Despite geopolitical predictions, it was not Russia or Iran – rather, neglected crucial technical, social, environmental and economic issues – that created obstacles for the project and increased the costs of it, and thus led to lengthy delays, the launching of investigations and almost, indeed, the halting of it entirely. During the construction phase, one of the BTC pipeline's key stakeholders, Georgia, blocked the project because of environmental issues. Its one of the western supporters, the UK, investigated and questioned the involvement of BP and ECGD in the project vis-à-vis technical and material failures. Finally, the project was delayed due to massive protests by workers, NGO pressures and land acquisition disputes. Contrary to state-centric Great Game assumptions, these issues were resolved due to the systematic support and networking of multiple players – namely, transnational energy companies, intergovernmental organisations and international banks. Considering the strong financial, political, technical and lobbying support they gave, it can be argued that the BTC would never have been completed without it.

The article has, finally, shown the neglected role of the BTC pipeline post-construction. Despite a number of geopolitical disputes and uncertainties, since its construction the pipeline has changed the dynamics of the Great Game in the region. More precisely, it has led to non-calculative cooperative behaviour and offered a functional way of pursuing pragmatic collaboration and exchange between the Caspian littoral states. The BTC infrastructure has become strategic because it offers the littoral states possibilities of contractual and material integration as well as adherence to international standards as alternative routes forwards to the unrealistic one of political integration, which helps them bypass thorny issues of nationalism, political difference and sovereignty while still creating material interdependence in the long run. In this regard, the BTC pipeline has created suitable conditions for pragmatic, flexible and technocratic cooperation – as the littoral states can be part of these activities with respect to their interests and resources while there is no obligation either to participate in the project or to stay out of it. Despite the long-standing issues existing between the littoral states, such as the uncertain legal status and contested nature of resource fields when it comes to Azerbaijan and Turkmenistan for example, the BTC pipeline has created room to circumvent these restrictions while still addressing issue-specific needs. Considering this, it can be argued that it is necessary to determine those activities which are common, where they are common and the extent to which they are common.

In terms of further research, this article has focused only on particularly salient issues in the three phases of the BTC project that others previously left unscrutinised. The downsides of infrastructural cooperation – for example, exploiting only specific natural resources or failing to develop the periphery – could not be discussed here. Another road forward comes from the generalisability of findings. The BTC pipeline is a specific case, one in which pragmatic and issue-specific cooperation can be observed. When using the same analytical structure (planning, construction and eventual use of infrastructure), scholars should now consider also regional differences. Overall, the research findings point towards a more cautious approach needing to be taken in evaluating the Caspian Sea region. They also complements a growing empirical literature in IR that deals with the role of infrastructure and technology in generating cooperation (e.g. Barry 2013; Latour 2005). More concretely, the findings of the article invite the IR literature to engage in a broader debate about the mutual relationship between infrastructure, technology and cooperation.

Acknowledgement

For helpful comments on earlier versions of the contribution, I would like to thank Tracey German and Philipp Olbrich as well as the two external reviewers.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Agha Bayramov is a PhD candidate and lecturer at the University of Groningen. His research interests are resource conflicts, energy security, the Caspian Sea region, and global governance.

Notes

1 The Baku–Novorossiysk pipeline is a 1,330-kilometre long oil one, running from Sangachal Terminal in Azerbaijan to Novorossiysk Terminal on the Black Sea coast, in Russia. The pipeline first became operational in 1997.

2 Used for coating girth welds, as well as valves, fittings, pipes, ballast tanks, ships and marine structures. SP-2888 is ideally suited for coating pipes to be used for slip bore/directional drilling, due to its superior abrasion, impact and gouge-resistance properties.

3 The Southern Gas Corridor is a term used to describe planned infrastructure projects aimed at improving the security and diversity of the EU's energy supply, by bringing natural gas from the Caspian Sea region to Europe.