Pipeline economics – why the Afghan pipeline will NOT be built

This topic comes up every now and then (and of course in the Michael Moore movie) and I try to shoot it down each times in a few lines, usually met with skepticism or mockery.

So here is the long version, once in for all, for future reference whenever this topic comes up.
Why it will not be built can be explained by having a detailed look at how pipelines are financed and paid for, and looking at how this applies to this project.

Just to be clear, the TAP (Trans-Afghan-Pipeline or Turmenistan-Afghanistan-Pakistan) is a proposed natural gas pipeline which would go from the gas fields of Turkmenistan to Pakistan through Afghanistan. All that follows below applies to both oil and gas pipelines, but I’ll focus on the gas case as it is what concerns us here.

A pipeline is very literally like a chain – all links must be in place for the whole chain to have any value at all. In the case of a pipeline, the links include a gas field to provide the throughput, the construction of the pipeline, the continued operation of the pipe, and a purchase of the gas on the other side.

What is essential to note is that to get any revenue from a pipeline, you need the whole chain to be in place – you cannot have two thirds of a pipeline, and you need the gas production and the gas consumption. That means that all the investments must come upfront and all the revenues will come only after all the spending has been made. As the tag price for a pipeline usually runs in billions of dollars (the typical price can be around 1-3 millions dollars per kilometer, depending on its size and the ground it covers), this means that financing such an investment is a fundamental question:

if you cannot say who is able AND willing to put 2 billion dollars on the table UPFRONT, and explain how they will get paid back, then your project will not fly.

Let me explain how this is usually done.

A pipeline is usually built by a gas producer who wants to gain access to the market or to a specific customer, by a customer needing access to gas reserves (think big customers like a power plant, a chemical factory or an aluminium producer), by a third party (usually, a specialised pipleine operator acting on behalf of the producer or the customer), or by any combination of the 3.

A gas producer wants to bring its gas to a market at the lowest cost possible. It has a good idea of how much gas he can produce and thus ship, and can determine a cost per unit of gas, which it can compare to the price it expects to sell the gas and its own cost of production. If the producer is reasonably confident to be able to sell its gas over the requisite duration (typically 15-20 years or more), it will invest in the pipeline, on a cost basis (i.e. the pipeline will effectively part of the cost of production of the gas from its perspective).

A gas purchaser is in a symetrical situation. It needs to connect to the site where gas is available (whether an individual gas field or a place where the gas grid already exists); it know how much gas it needs over the life of its industrial asset (again, 20-30 year periods are fairly standard) and the cost that this adds to the purchase price of gas over the long term.

A third party will build a pipeline if it can profit from it, as it is not involved in either gas production or consumption and cannot make a profit from the rest of the chain. It is possible to build “merchant” projects, i.e. “build it and they will come” – you build the infrasturcture and charge for its use. This is possible only in places where there is a lot of suply and a lot of demand and not enough transport capacity, which does not happen very often. In most cases, the third party is a pipeline operator acting on behalf of the gas producer or consumer, and the ownership is shared between them in various combinations (everything is possible); the only important thing in that configuration is that the pipeline is an independent entity which must make a profit.

In that situation, there are several ways to remunerate the pipeline company:

• a simple tariff, proportional to the volume of gas shipped
• a “capacity” charge: i.e. the user pays for the right to use a given fraction of the pipeline capacity, whether it actually uses this capacity or not
• or any combination of the two.

A typical situation is a capacity charge which is high enough to guarantee a minimum level of revenues (ideally, enough to pay off the initial investment on its own), and a low tariff which reflects operational costs for the use of the pipeline and provide potential profit for the pipeline operator (a minimum level of use will provide a small profit, a full utilisation will yield a nicer, but never extravagant, profit).

Another way to materialise such arrangements are “ship-or-pay” contracts, whereby there is only a tariff proportional to the volume, but with an obligation to pay it anyway, up to a certain value, even if the corresponding volume is not shipped (the shipper then getting “make up” rights – i.e. it can ship more without paying for it again if it exceeds the requisite volumes in future periods.

The essence of all these arrangements is that someone has to commit to provide a minimum level of revenues to the pipeline operations in order to pay off the initial capital investment. Such commitment is what makes a project economic and usually makes it financeable as well.

For someone to commit to paying such tariff – and remember, a pipeline usually requires 15 years of operations for the tariff to make economic sense – it has to have a pretty good certainty that (i) it will need the capacity for such a period, ((ii) it will have use for it and (iii) it will be able to afford it. Such a commitment to pay can be a major financial drain if the corresponding revenues (from selling the gas or from using it) are not there.

So we’re back to our initial questions, but with more details:

• are there enough gas reserves to fill up the pipeline capacity for the requisite 15-20 years?
• is the gas producer able to produce the requisite volume for 15 years? (has he invested enough to produce the gas?; is the production profile compatible with the transport infrastructure? are all the permits, authorisations, etc… necessary to exploit the gas fields available, and can they be expected to remain in place? do the production costs – including all taxes – make sense in view of the whole chain?)
• is the gas producer committed to delivering these volumes through ths pipeline?
• are the proposed construction costs for the pipeline realistic, and will the construction schedule be met?
• is the pipeline operator experienced and able to keep it functioning for the required duration at the required capacity?
• has the pipeline obtained all the necessary permits, licences, authorisations from all relevant authorities?
• will there be a market or a buyer to take all the gas for the requisite 15-20 years?
• are the purchasers able to pay for the gas for the period?

which can also be identified as follows:

• reserve and production risk
• producer commitment risk
• construction and operation risk
• market and price risk
• political risk
• buyers’ counterparty risk

ALL these risks must be acceptable for the project to make sense. Any major issue in any of these categories is sufficient to kill the project. Banks and investors look at it the same way, with the simple difference that, as banks’ revenues are imited at most to the interrst income, theyt alos want to limit their risk. As a result, they usually get a first dip in the revenue, after operating costs but before investor revenues.

So, what about our Afghani project? Let’s look at all the above points in turn:

– gas reserves and production

That’s clearly the strong point of such a project: Turkmenistan has massive gas reserves (the fourth in the world) and it already has significant production capacity (including inutilised capacity since the break up of the Soviet Union). So the requisite gas is there and could be produced and shipped in the required volumes.

– gas delivery commitment

Unfortunately, this is the biggest hurdle for the project: you need to trust the Turkmens to deliver their gas to the pipeline for the next 15 years. The risk is especially important as Turkmenistan is the only possible source of gas for the pipeline and their continued participation in the scheme is therefore essential. The risk is two-fold:

• the political risk is extremely high, with Turkmenistan ruled by a crazy dictator with absolute powers. He has shown that he was not necessarily rational and could change his mind very easily; if he does that about the project, there is no recourse. Being a dictator, should he fall, it is not clear that his successors would honor a commitment that he made. Over 15 years, these risks are significant.
• the second item, and more important one, is that Turkmenistan already has an available route to export its gas via the pipelines going North to Russia. These pipelines have been built a while ago (during Soviet times) and do not have to be paid for anymore. They are thus available immediately, and at a very low cost (operating costs, which are usually low for pipelines). That means that it is quite easy for the buyer of gas at the end of these pipelines (currently, the Russian monopoly Gazprom) to offer at any time a higher net price for Turkmen gas than they can get on the other side.

The fact that the Afghan pipeline would not be competitive is thus a major obstacle to its economic rationality, as it threatens the availability of the Turkmen gas volumes.

– construction and operations

This is not an dealkiller, as pipelines have been built in many difficult or harsh places, but it is clearly a challenge. Building a pipeline requires bringing massive quantities of steel (count a few hundred tons per kilometer) – and the workers to put them in place to locations out of reach of roads and other transportation modes. Afghanistan has few roads, a harsh climate, and it would thus be a complex logistical exercise. The risks are thus both high as regards the cost of construction and its time schedule. and any delay has major economic implications as interest costs run on the full amount of the initial investment and are compounded as delays mount.

– market and price risk

The proposed market for the gas to be shipped is the Pakistani market, and possibly (but after additional investments are made), the Indian market (requiring a pipeline between the two countries) or the international market (requiring the construction of a liquefaction plant on the Pakistani coast). The Pakistani market is likely to grow over the coming years, but it is a hard market to assess. In any case, the pipeline company would not want to distribute the gas itself and would thus rely on a local counterparty, in all lielihood the national gas company (Pakistan Petroleum Ltd, PPL). The project thus requires this company to commit to take the requisite volumes for the requisite period, and to pay for it over the duration – in hard currency. This is a risk that the banking market will NOT bear and that international oil & gas companies are unlikely to take themselves except if they have a natural hedge through local production, which is incompatible with a pipeline import project. Multilateral institutions like the Asian Development Bank or the World Bank might be able to do it, as well as export credit agencies (government agencies from the rich countries which subsidise exports from their countries by guaranteeing payment risk), but they usually require commercial banks to share a part of the risk in such big projects.

The recent experiences of Dabhol  (a big power plant in India) and Argentina further show that ven if the demand is there and the price (in dollar terms) is guaranteed by a public body, the commitment to pay these amounts in situations when there is a currency devaluation but no significant increase of domestic prices for gas or electricity is very weak, and investors end up being paid in worthless local currency – starkly insufficient to repay dollar debt.

– political risk

This is also a major obstacle. This is a 3-country project, and these are extremely rare. As far as I know, the BTC pipeline from Azerbaijan to Turkey via Georgia is the only recent example, and it’s taken the combined might of BP and a dozen other oil majors with 5 billion barrels of oil on their hands and no other way to bring them to the market, the full support of the US government (fighting against Russia and Iran), the presence of the World Bank, the EBRD and 6 Western government export credit agencies to pull it through – and it’s taken 10 years.

In this case, you can argue that you probably have the worst combination imaginable – a crazy dictator, a country with warring local warlords and almost no centralised government, and a highly unstable country – and you need each of them to be happy at all times for a full 15 years, not renege on ANY of their commitments, and not try at any time to get a better deal (with each being absolutely indispensable to the project). Hard to imagine, even with 15,000 US soldiers on the ground…

– counterparty risk

as all counterparties in the 3 countries are public entities, this is fairly similar here to political risk with the price risk added in Pakistan. There are no majors involved in gas production in Turkmenistan, and none in gas consumption in pakistan, so you rely in each case on the local actors. The pipeline would likely be built by a consortium including an oil major, and you could expect that part to be at least manageable, but that’s not enough.

So, you’re going to tell me, if this project is as impossible as I claim, why do we keep hearing about it? And why do we find these suspicious connections between senior political figures in Afghanistan and oil companies?

Fair questions, but with relatively simple to answer in fact.

The 3 countries would like this project to exist. Turkmenistan would like to have an alternative to Russia to sell its gas to, Afghanistan would like the transit revenues it would bring, and Pakistan does need gas and this is one of the options. A lot of people are going to tell the authorities of these countries the things they want to hear, i;e. that this project can be built in a painless way. Some institutions may have other interests (the ADB would like to show that it can do a major oil&gas project, some of the oil producers have operations in Pakistan that they may want to protect or expand, and various countries in and out of the region have various interests involved and want to support their allies and their pet projects). The question, as stated above is – who will put 2 billion dollars upfront in this project? Putting a few million to conduct feasibility studies, naming a roving ambassador that makes speeches, etc… costs nothing to an oil major or a big country, and brings various diplomatic or relationship advantages, but it does not finance or build a project.

So, please, please, do not use the Afghan pipeline as an axample of nasty oilmen conspiracies. There are enough of these going on not to focus on those that have no serious basis in reality. It just makes you lose credibility with those that know anything about the sector.

Remember, oil is a mutli-hundred-billion dollar business. Spending a few million here or there to make or keep friends and make them believe you are their friend is a small investment in the larger scheme of things. Making big announcements is a way of life for politicians and it costs oil companies to flatter them by letting them having their ways and the positive PR even if there is nothing behind the announcements.

It’s not because Halliburton does evil stuff (mostly scamming the US government by the way) that everything that any oil company does is evil or suspicious…

Please bring up your questions or suspicious quotes and I will try to answer them as best as I can.