Years of planning and months of exhaustive work by thousands of advisors, investors and developers came to a head on 3 February 2004 with the financial close of the one million barrels a day 1,780km Baku-Tbilisi-Ceyhan (BTC) pipeline, writes Max Thompson.

This was an outstanding project that deserves to win Infrastructure Journal’s Project of the Year 2004 award on a number of different levels, not least being that it took hosts of players a few months to nail down a transaction that stretched over three countries and had to create groundbreaking political, legal and financial frameworks.

In addition to breaking into a virgin market, the financing was closed successfully in less than three months in what could only be described as a turbulent political climate.

The project followed hot on the heels of what became known in November 2003 as the Rose Revolution in Georgia, a key transit country, and the death of Haidar Aliyev, resulting in a change of government in Azerbaijan the very next month.

BTC and the BTC sponsors worked closely with the new governments to provide the lenders assurance of continued political support for the projects in both countries.

This resulted in the launch of a path-breaking environmental and social programme. The BTC and ACG projects together were the largest private infrastructure projects in the world in 2004 – and the BTC project in particular came under a tremendous amount of scrutiny by NGOs, governments and others, especially with respect to environmental and social matters.

BTC and its sponsors worked closely with their advisors to develop a package of environmental and social documentation that provided the lenders comfort as to the environmental and social performance of the project.

It was also the first major project to be developed following adoption of the Equator Principles, and successfully worked with the commercial bank lenders to determine how these principles would be applied in practice and demonstrated to their satisfaction that its requirements had been met.

The closing of the US$3.6bn project represents a genuine milestone in project finance history not only because of the huge geo-political scale of the project, but also because of the legal and financial complexity generated by the large number of stakeholders involved in the scheme.

The fact that the last of some 17,000 signatures was inked at the signing ceremony in Baku is testament to the volume, and diversity, of the parties involved.

When it becomes operational later this year the pipeline – which is being developed by the BTC Pipeline Company (BTC Co) – will provide an alternative route to European markets for oil from the resource-rich Azerbaijan. And the pipeline is being fed by the BP-operated Azeri, Chirag, Gunashli (ACG) oilfield which is located offshore Azerbaijan in the Caspian Sea.

The pipeline – which is slated to have an operational life of 40 years – runs north-westerly from Baku in Azerbaijan to Tbilisi in Georgia from where it stretches across Turkey to the Mediterranean port of Ceyhan, in the south-west of the country, and in doing so crossing 2,420 bridges, roads and rivers.

As John Wingate, finance manager of the BTC Co, says: ‘With more than 99 per cent of the line pipe now welded and the end of construction in sight, the BTC pipeline is close to becoming an actuality.

‘This project would not have been possible without the support of countless people and institutions, it is a tribute to all those who have worked so hard over many years to craft what will become one of the most significant infrastructure projects of recent times.’

BTC Co is an international consortium formed of SOCAR – the state-owned oil company of Azerbaijan – BP; Turkish firm TPAO; Statoil; Unocal; Itochu; Amerada Hess; Eni; TotalFinaElf; INPEX and ConocoPhillips.

The project also includes the Sangachal terminal in Baku – which has two storage tanks – and the Ceyhan export terminal in Turkey which has seven crude oil storage tanks and a jetty with the capacity to load two tankers simultaneously. In addition, eight pumping stations will be built – two in Azerbaijan, two in Georgia and four in Turkey.

Background and Rationale

When first conceived under the so-called Early Oil Project, oil from the ACG field had two proposed export routes – the Northern Route Export Pipeline (NREP) that runs from Baku up the north-east Azerbaijan coast across Russia to the eastern edge of the Black Sea; and the other option was the Western Route Export Pipeline (WREP) that runs through Georgia to the south-eastern coast of the Black Sea.

However, both of these routes then require shipments across the Black Sea, eventually making it to market via the Turkish Straits.

In 1994 when the ACG partners signed their production sharing agreement, they acknowledged that with only the NREP and WREP as export routes, a greater security of supply was required and in 1997 started looking for a third export route.

However, the most direct export route to the Mediterranean entails the traversing of the volatile and politically sensitive countries of Iran and Armenia – a risk too far for many investors.

A Complex Project

The BTC was not an easy pipeline to build. As well as the thousands of rivers and bridges the crude oil within the pipe will cross, during its long journey – 465km in Azerbaijan, 255km in Georgia and just over 1,000km in Turkey – it will also tackle a 2,500m Turkish mountain range.

Its technical challenges were also compounded by security risks, particularly on the territories of Georgia and Turkish Kurdistan. However, most troubling for developers during the initial stages of the project was BP’s opposition to the project on the basis that not enough oil had been found to justify the high project costs.

Fortunately for the project BP in 1999 revised downwards the amount of oil reserves that would be needed to make the pipeline economical from 6 billion barrels to a more achievable 4 billion to 4.5 billion barrels.

Seven of the 10 members of the original Azerbaijan International Operating Company (AIOC) consortium – the AIOC was formed in 1994 to develop Azeri oil potential – decided to go ahead with the BTC sponsor group, with only Lukoil, ExxonMobil, and Devon Energy declining the offer.

Meanwhile, SOCAR, which originally had a 50 per cent stake in the sponsor group, sold ENI – a non-member of AIOC – a 5 per cent share in the pipeline project in October 2001. After failing to come to agreement with other energy companies to join the sponsor group, in March 2002 SOCAR reduced its stake in the pipeline project to 25 per cent, distributing 20 per cent among other group members. In June 2002, SOCAR sold an additional 5 per cent share to TotalFinaElf, but rejected a proposal from ChevronTexaco to join the sponsor group.

At the end of June 2002, the head of the sponsorship group, Michael Townshend of BP, said that the pipeline ownership group was complete.

The difficult task of convincing developers to get involved required solving two sets of issues: first, drafting an agreement covering the building and operation of the pipeline which sufficiently ring-fenced the potential legal, regulatory, security and tax problems that could arise between Azeri, Georgian and Turkish institutions. That was finalised at the Istanbul Summit in November 1999.

Secondly, project financing of the project had to be secured – the complexity of which many believed would prove too arduous a task to overcome. However, the lending group and sponsors did ultimately pull together to finalise what was to become a spectacular deal.

Legal hurdles

The legal complexities inherent to the project were compounded by the number of jurisdictions the pipeline was to traverse. It was only by having draft contracts that determined the pipeline’s legal foundation that the companies could arrive at the actual cost.

To achieve this, the companies demanded that each host government provide predictable construction costs, transit fees, taxes, land use rights, and physical pipeline security.

The myriad of different domestic laws was a constant headache for the parties, helped not in the least by antiquated Azerbaijan and Georgian legal systems that were still in transition and had not yet altered laws prohibiting foreign companies from owning land. So, instead of having local laws amended, a binding international treaty was created that would supersede the domestic laws of each of the three host countries.

IOCs would also not trust local courts to enforce the terms of the agreement. So, the consortium insisted that all conflicts arising under the contracts be submitted to private arbitration under UK law.

While such agreements are common for oil and gas projects, this time – instead of one contract between one consortium and one government – separate agreements were struck with each government individually, and another one between the group of companies and all three different governments collectively.

The agreements do not only supersede current domestic laws, but also all future laws—for up to 60 years. Any subsequent environmental regulation cannot apply to the consortium project. The same goes if one of the host countries wants to raise corporate taxes, or should it end up involved in local conflict and the government decides that it needs more money for pipeline security.

Furthermore, Azerbaijan, Georgia, and Turkey each had unstable tax laws with rates higher than the consortium was willing to pay. A tax structure which – by keeping taxes associated with the project as low as possible – would keep down the cost of building and running the pipeline. Each country was made to agree to exempt the contractors and subcontractors building and operating the pipeline from domestic taxes, thereby dramatically lowering costs for the consortium.

Security was also an issue. Because pipelines are frequently targets of sabotage or terrorism, the companies insisted that each country commit its security forces to ensuring the project’s safety – during its construction and afterwards.

The contract exempts the consortium from any legal responsibility for the actions of those security forces.

Land acquisition was another sticking point. Although associated conflicts with landowners were left to local courts to decide, the companies got what they wanted – a guarantee of complete access to the land.

Finally, in 1999, the IOCs realised they had yet another serious problem to overcome; the Turkish portion of the pipeline was projected to cost more than the consortium was willing to pay. So Turkey made an offer – the country’s state-owned pipeline company would construct and operate the Turkish portion of the pipeline for far less than the oil companies’ estimates suggested. Although the companies would lose some control over the process, Turkey’s promise to build it for a fixed cost was necessary to save the project.

Financing

Having dealt with political and legal pitfalls, in late 2001, negotiations with the European Bank for Reconstruction and Development (EBRD) and the International Finance Corporation (IFC) commenced on the premise that the two multinational lending agencies would form the core of the lender group.

Swift on the heels of the IFC/EBRD negotiations, came the appointment of a number of export credit agencies (ECA) which by mid-2002 included the Japan Bank for International Cooperation (JBIC), the Nippon Export and Investment Insurance (NEXI), the Export-Import Bank of the USA, the UK’s Export Credits Guarantee Department (ECGD), Compagnie Francaise pour le Commerce Exterieur (COFACE) of France, Euler Hermes Kreditversicherungs (HERMES) of Germany and Servizi Assicurativi del Commercio Estero (SACE) of Italy.

ABN AMRO and SG – which previously had been awarded short-term mandates to undertake a bankability study into the BTC project – were joined by Citibank and Japanese bank Mizuho and all were appointed as initial mandated arrangers. The four banks worked alongside EBRD and IFC to structure a bankable deal and formulate a comprehensive terms sheet.

Each of the four banks had originally been assigned specific roles – for example, ABN AMRO was designated coordinating bank to the European ECAs – but in summer 2003, the banks’ separate roles were consolidated into a formal arranging role and the initial mandated leader group was formed in which the banks were also joint book runners.

Twenty banks were then approached, of which 15 provided commitments of around US$1bn. The commercial banks were led by ABN AMRO, Citibank, Mizuho and SG and joined by Banca Intesa, BNP Paribas, Crédit Agricole Indosuez, Dexia, HypoVereinsBank, ING, KBC, Natexis Banques Populaires, San Paol-no IMI, WestLB and Royal Bank of Scotland.

It was soon agreed that there was not a need for a general syndication, and each of the 15 banks was appointed as a mandated lead arranger.

Talking of the strategy, Stephen Bedwood, director of structured capital markets at ABN AMRO, says: ‘The syndication strategy adopted from that stage was to go with an expanded arranger group rather than a conventional underwriting with a large retail syndication.

‘Initially we didn’t expect the group to be so large and considered the possibility of having two tiers formed of a mandated lead arranger group and a co-arranger group. But the success of the syndication at that top level meant that we didn’t have to go forward to a retail syndication phase.’

In addition to the ECAs, commercial banks and the main lending agencies, there were two further groups that contributed to the project: The US-based Overseas Private Investment Corporation (OPIC) provided political risk cover in a separate tranche of commercial bank debt and, in addition to providing an export credit facility, JBIC also provided an untied loan of US$300m.

Conclusions

Looking at the BTC pipeline project and the work that went in to linking the oil fields of Azerbaijan with the resource-hungry European markets, it is a feat of political, legal and financial genius to have closed such a massive transaction the crossed three borders – borders that at times have been far from friendly.

The project fulfills and surpasses all criteria to be voted by Infrastructure Journal’s panel of judges to be voted IJ Deal of the Year 2004. It is a feat that is all the more impressive in that it beat international rivals – all of which had a legitimate claim to the title.

BTC stands as testament to the efforts of all the players who started investigating the potential of shipping oil from land-locked Azerbaijan passing through Georgia and terminating in Turkey, on the Mediterranean and then on to the Western markets.

Prior to the BTC project, the largest foreign investment to the region through which the pipeline snakes had been measured in tens of millions. This underlines the impact on the local countries the significance of the projects to the impacted countries, and the reflected glory and kudos for all the professionals involved in making it a reality.

In addition to the size of the debt, the claim that the BTC documentation contains 10 signatures for every kilometre of pipeline is a telling indication as to the meandering complexity of the cross-border transaction.

As Bedwood says: ‘There is no getting away from the fact that the deal was a tremendously complicated transaction. When one considers the political risk, the size of the sponsor group and the different types of institutions – inevitably challenges and complexities had to be dealt with along the way. However, the syndication has been extremely well received and, the ultimate stamp of approval is that the market views this as a well structured deal.’