Baku-Ceyhan pipeline: bad economics, bad politics, bad idea

COMMENT On Baku-Ceyhan oil pipeline

Joe Barnes, Ronald Soligo
Houston

Central Asia/transcaucasus oil Pipelines [232,985 bytes]

Much of the research underlying this article was sponsored by the Baker Institute for Public Policy and funded by the Center for International Political Economy.

It now appears that the long-awaited decision on a main export route for oil transported from the Caucasus and Central Asia will be against immediate construction of a much-discussed multibillion-dollar pipeline from Baku in Azerbaijan to Ceyhan on Turkey's Mediterranean coast.

Press reports suggest that interested parties have failed to develop a commercially viable plan for the Baku-Ceyhan pipeline (OGJ, Oct. 19, 1998, Newsletter), and are leaning, for the foreseeable future, toward more modest routes from Baku to the Georgian port of Supsa and/or the Russian port of Novorossiisk. The delay marks a sharp reversal for the Clinton administration, which has made the early construction of a Baku-Ceyhan pipeline a centerpiece of its policy towards Turkey, the Caucasus, and Central Asia. But it also reflects good economic and political sense.

The stakes are admittedly high. Today a modest producer, the Caucasus and Central Asia could become a significant supplier of oil to world markets by the end of the next decade. Total output could rise to 3-4 million b/d by 2010, making the Caucasus and Central Asia a producing area roughly as important as the North Sea is today.¹

But getting the region's petroleum to consumers will be difficult. The countries of the Caucasus and Central Asia are landlocked, impoverished, buffeted by internal instability, and the object of intense rivalry by external powers such as Russia, Iran, and, increasingly, the U.S. While Kazakhstan has by far the greater hydrocarbon resource potential, moving expanded Azerbaijani oil production to market is an important first test of the region's viability as a major petroleum producer.

Transportation hurdles

Identifying a major new export route-much less financing or constructing one-has proven to be difficult.

There has been no shortage of proposals and no shortage of objections-economic, political, or both-to each of them. Recent discussion has focused on western routes that would bring Azerbaijani oil to the Mediterranean and from there to European markets via a combination of pipelines and tanker transport.

The reason for past delay is not hard to find. Literally billions of dollars are at stake for the government of Azerbaijan, international energy companies, and transit countries. The choice of route will determine not just the overall price tag of the project but also how its costs and benefits will be divided among the various parties. As always, "Who pays?" and "Who gains?" are as important as "How much?"

The Clinton administration has made its own preference clear. It has supported the construction of a 2,000- km pipeline from Baku in Azerbaijan, through Georgia and then to Ceyhan on Turkey's Mediterranean coast, where the oil would be loaded onto tankers for onward shipment.²

But the Baku-Ceyhan pipeline has its own problems, not least its cost: at an estimated cost of $3-4 billion, it is by far the most expensive of the western options.

Assuming a capacity of 800,000 b/d, Baku-Ceyhan would result in total transport costs of roughly $2.80/bbl from Baku to Italian ports. This figure includes capital, construction, operation, and all sea freight costs.3 It excludes, however, any fees that might be levied by Turkey and Georgia, the two transit countries.

Baku-Ceyhan alternatives

There are other, less expensive western alternatives (see map, p. 29). Most involve expanded shipment of Azerbaijani oil by pipeline to two northern Black Sea ports, either Novorossiisk in Russia or Supsa in Georgia. There is an existing pipeline from Baku to Novorossiisk that could be upgraded if necessary. A refurbished line to Supsa will begin exports by April 1999.

From Novorossiisk and/or Supsa, Azerbaijani oil would then be shipped to ports on the western and/or southern shore of the Black Sea for onward shipment to the Mediterranean by pipeline.

One proposal would involve moving the oil to Samsun on the Turkish Black Sea coast and then to Ceyhan, shortening the length of the necessary pipeline dramatically.

Another proposal would involve shipping the oil to Burgas in Bulgaria and then by pipeline to the Aegean port of Alexandroupolis in Greece.

Yet another option is a pipeline across Thrace from Kiyikoy on the Black Sea to Ibrikbana on the Aegean. The route is both relatively short-only 190 km-and entirely within one country, Turkey. Using it would admittedly be more complicated than simply shipping Azerbaijani oil via Baku-Ceyhan. It would require shipping Azerbaijani oil by pipeline to Novorossiisk or Supsa, moving it by tanker to Kiyikoy, offloading it there, and loading it onto a tanker again at Ibrikbana.

Cost differences

Even so, the total transport cost to Italy of the Supsa route-including construction of an 800,000 b/d pipeline across Thrace, expansion of the Baku-Supsa pipeline, and costs associated with loading and offloading-is roughly $1.90/bbl, $0.90 less than the comparable figure for Baku-Ceyhan.This figure again excludes any transit fees Georgia and Turkey might charge.

The advantage over Baku-Ceyhan, already significant, would be even greater were a larger Thracian pipeline built. This would have the advantage of accommodating Russian as well as Azerbaijani exports. A 1.5 million b/d Kiyikoy-Ibrikbana pipeline, for in-stance, would have a $1/bbl advantage over Baku-Ceyhan.

There exists another and still cheaper alternative: simply moving additional Azerbaijani oil via pipeline to Novorossiisk and/or Supsa and then by tanker through the Bosporus Strait to the Mediterranean. This is a route Azerbaijan already uses, although at relatively low volumes. Supsa is an attractive option. At $1.40/bbl, the cost of shipping from Baku to Italian ports via an expanded Baku-Supsa pipeline and the Bosporus is $0.50 lower than Kiyikoy-Ibrikbana and $1.40 lower than Baku-Ceyhan. Again, any transit fee by Georgia is excluded.

These transportation cost differentials are significant. In the case of Baku-Ceyhan, they are about $1.1 million/ day, assuming 800,000 b/d of exports. For a Thracian pipeline, they are about $400,000/day with the same assumptions.

Who will pay such a differential? Not the European consumers of Azerbaijan's oil. Today's world markets are competitive and set one price for petroleum of similar quality. This leaves sellers: the producing companies or Azerbaijan. Private oil companies that have invested in Azerbaijan-notably those belonging to the Azerbaijan International Operating Co. (AIOC), an international consortium that includes British Petroleum Co. plc, Amoco Corp., Exxon Corp., and Pennzoil Co.-will absorb much of the cost.

But so, undoubtedly, will Azerbaijan itself, which sees the prompt development and export of its oil reserves as a high national priority.

For sellers, using Baku-Ceyhan could lead to a revenue loss of about $1 billion/year. Roughly $400 million of the foregone revenue represents the additional cost of shipping oil through Baku-Ceyhan instead of using the Bosporus. Much of this is what economists call a "deadweight loss." It is an expenditure by Azerbaijan and oil companies that might otherwise have been better used by themselves-or even, for that matter, by Turkey.

The latter point becomes clear by asking a simple question: would Turkey prefer $400 million/year spent on the construction of an inefficient pipeline or $400 million/year in cash to spend on something else? The $400 million figure, significantly, excludes any transit fees that Georgia or Turkey may impose. These could dramatically increase the cost to Azerbaijan and producing companies. A modest $2/bbl tariff on oil moving through Baku-Ceyhan (in addition to pipeline costs) would cost $585 million in additional lost revenue per year.⁴

Whatever the share of the total billion-dollar loss that is borne by Azerbaijan, it will surely be significant for a country whose 1996 GDP was only $3.2 billion.

Bad deal

The bottom line is clear: any Turkish pipeline is a bad deal for Azerbaijan and the oil companies-and Baku-Ceyhan a very bad one, indeed.

Then why has the Clinton administration pushed it? One reason is clearly deference to the Turkish government's concerns about congestion and pollution in the Bosporus Strait.

The Turks have cause for worry. After all, the narrow strait twists and turns through the center of greater Istanbul, a highly urbanized area of 12 million inhabitants. And it is a very busy waterway, indeed: on average, 136 vessels transit the strait daily, some carrying potentially hazardous cargo. This is in addition to the 2,000 or so trips by local ferries and other boats from one side of the Bosporus to the other. The risk of accident is significant. In 1994, for instance, an oil tanker collided with a freighter and burst into flames, killing 30.

Moreover, the current legal status of the Bosporus-which is, under the Montreaux Convention of 1936, an international waterway with very limited supervisory authority granted Tur- key-makes it impossible for Ankara to exact tolls that could relieve congestion and impose insurance requirements that could offset environmental costs.⁵

But it is far from certain that Turkey will, in fact, gain much benefit in terms of congestion or pollution abatement by forcing Azerbaijani oil to bypass the Bosporus. The 800,000 b/d that would be diverted through Baku-Ceyhan would represent, after all, only one large tanker. Improved tracking systems or more careful scheduling would go far towards reducing this additional congestion. The lessened risk of environmental damage would appear minimal: other oil tankers, notably from Russia, as well as vessels carrying even more volatile and dangerous cargoes, would continue to navigate the strait.

Turkey's benefits

Turkey will likely gain significant revenues from transit fees associated with any pipeline through its territory. This has led many to suspect that Turkey's concerns about the Bosporus, although real, have been exaggerated as an argument in favor of a pipeline that would benefit Ankara financially.

Baku-Ceyhan, given its length, could be particularly lucrative from a Turkish point of view. But it is important to recall who will bear this cost: the producing firms and Azerbaijan. The U.S. is a long-time ally of Turkey, a NATO (North Atlantic Treaty Organization) member. As such, we would presumably welcome any additional revenue that Turkey might gain from a Baku-Ceyhan pipeline.

But the U.S. has also supported the development of Azerbaijani oil, at least in part, as a measure to promote that country's economic welfare and political independence. It makes little sense, therefore, to advance a pipeline project that represents a major transfer of revenue from Azerbaijan to its richer and more stable neighbor to the south.

Iran, Russia concerns

Washington also clearly sees Baku-Ceyhan as an alternative to a pipeline through Iran-something that the U.S. has long and vehemently opposed. ⁶ But any of the western options-whether Kiyikoy-Ibrikbana, another bypass, or simply moving additional tankers through the Bosporus-can get Azerbaijani oil to market without transiting Iran.

Less publicly, some in the Clinton administration also believe that building Baku-Ceyhan can help insulate Azerbaijan from potential pressure by Moscow. Relations between the two countries have been troubled since Azerbaijan achieved independence; Moscow clearly resents what it considers Baku's insufficient deference to Russia's traditional leadership role in the Caucasus.

But Russia, if it wants, can slow or stop the flow of Azerbaijani oil that uses any Western route currently under consideration. It can do so directly with the Baku-Novorossiisk pipeline. And it can do so indirectly with any route that transits Georgia, including Baku-Ceyhan. Russia retains a significant military presence in Georgia and has destabilized that country in the past.

Whatever its reasons-and none even approach compelling-the Clinton administration has put itself in the position of making Azerbaijan and the shareholders of oil companies pay for U.S. foreign policy. It could reduce, if not eliminate, the cost they bear by subsidizing construction of the Baku-Ceyhan pipeline. And, indeed, the Clinton administration has offered the backing of the Export-Import Bank and the Overseas Private Investment Corp. (OPIC) to the Baku-Ceyhan project. But such subsidies are unlikely to rise to a level necessary even to begin compensating Azerbaijan or oil companies for the huge losses associated with using a Baku-Ceyhan route. Amoco has suggested that those subsidies would need to be $365-550 million/year; our own calculations, as we have shown, suggest an even higher figure.⁷

How, then, can the administration claim that Baku-Ceyhan is "commercially viable" with Ex-Im Bank and OPIC support? By pointing out that Azerbaijan and oil companies will still turn a profit. This is true as far as it goes-but conveniently overlooks the huge additional profits that both will forgo under Baku-Ceyhan.

Infeasible option

Baku-Ceyhan is a pipeline that makes little sense from either an economic or, for that matter, political point of view. The day may come when congestion in the Bosporus-and the costs of delay-might make it economically rational to construct a pipeline across Thrace. But that day clearly has not yet arrived. The Russian government, for instance, has estimated that the costs of delay at the Bosporus for 349 Russian-flagged vessels during 1994-97 was less than $2,700/ship-a nearly trivial amount by any standard. ⁸

Assuming a demurrage cost of $60,000/day for a large tanker, the average delay in transiting the Bosporus would have to rise to roughly 6-7 days before it made economic sense to use a Thracian pipeline instead of using the strait.

The day when a Baku-Ceyhan pipeline makes sense is more distant yet. Average delays would have to rise to perhaps 18-19 days before using it becomes preferable to transiting the Bosporus.

Even if the Montreaux Convention were renegotiated and Turkey granted the right to levy tolls on ships moving through the Bosporus, it would likely still be more economic to use the strait. Assuming transit fees of $2/bbl for Baku-Ceyhan, the Bosporus toll would have to be $2.7 million for a tanker carrying 800,000 b/d of oil before Baku-Ceyhan becomes competitive.⁹ Assuming a $0.50/bbl transit fee for a Thracian pipeline, the Bosporus toll would have to be $800,000 before using the pipeline becomes economic.

Conclusion

In short, Baku-Ceyhan is an idea whose time has not come and, indeed, may never come-despite Clinton administration suggestions that the likely decision to use a Supsa route is only a delay, not a final rejection. ¹⁰

It may be a worthy economic and political goal of the U.S. government to facilitate the oil exports of Azerbaijan and to address Turkish concerns about the Bosporus. But a Baku-Ceyhan pipeline does neither.

For now, it is clearly better to move additional Azerbaijani exports through the Bosporus. Turkey's concerns about the strait-which are, indeed, legitimate-are best addressed through multiparty talks aimed at developing some rational system for use of that waterway. The U.S., as the world's foremost maritime power, could and should play an important role in any such talks. But to push Baku-Ceyhan-economically foolish in extreme and politically dubious at best-has never made sense and still does not.

References

2. Testimony of Undersecretary of State Stuart Eizenstat before the Senate appropriations subcommittee on foreign operations, May 31, 1998; and remarks by Energy Sec. Federico Pe?a at the U.S.-Azerbaijan Chamber of Commerce, May 21, 1998.

3. Throughout this paper, all transportation estimates include these costs. For a more comprehensive discussion of pipelines, see "The Economics of Pipeline Routes," Ronald Soligo and Amy Jaffe, The Baker Institute, 1998; and "Energy Choices in the Near Abroad: The Haves and Have-Nots Face the Future," Robert E. Ebel, CSIS, 1997.

4. By comparison, Russia currently charges a $2.16/bbl transit fee on the 120,000 b/d of crude oil transported through the 1,600 km Baku-Novorossiisk pipeline. Georgia will charge a $0.40/bbl transit fee on the renovated 200,000 b/d, 850-km Baku-Supsa pipeline, which is slated to come on stream early in 1999.

5. In early September 1998, the government of Turkey announced its intention to impose an insurance obligation on vessels passing through the strait.

6. In her May 18, 1998, statement lifting the threat of sanctions against European and Asian firms seeking to invest in Iran, U.S. Sec. of State Madeleine Albright reiterated the Clinton administration's opposition to a trans-Iranian pipeline.

7. Amoco was quoted in the Houston Chronicle, Nov. 28, 1997. Lee Hamilton (D-Ind.), ranking minority member on the House international relations committee, is on record recently opposing any subsidies. See The Journal of Commerce, July 14, 1998. In any case, BP's recently announced agreement to purchase Amoco has made U.S. subsidies for a project now dominated by a non-U.S. company much more problematic.

8. "Ship Reporting and Related Matters," a memo by the Russian Federation to the International Maritime Organization's Subcommittee on Safety of Navigation, Apr. 10, 1997.

9. The current toll at the Suez Canal for a similar vessel is $150,000.

10. Richard Morningstar, White House special adviser on Caspian Basin energy policy, quoted in the Houston Chronicle, Oct. 13, 1998.