Function of the Power Purchase Agreement (PPA)?
A PPA is designed to allocate risks between the parties. There are various categories of risks, including risks which are not within the direct control of either the project or the government. Unless these risks are specifically identified and addressed, there is a strong possibility that the force majeure clause will be triggered, resulting in a suspension of the agreement although not its termination. But the effect is that the generator is not paid.
The project agreements are structured with the aim of moving risks to the party best able to bear the given risk. In the case of risks outside the direct control of the parties, clearly no one is in a particularly strong position to manage the risk, but the government will usually be in a less weak position than the project company if for no better reason than that it is better able to absorb potentially large losses.
Risks addressed in the PPA
Hydrological risks are obviously only found in hydro projects – the nearest equivalent in thermal power is the location risk. Location risk is relatively easy to deal with, given that a thermal power plant can be sited virtually anywhere within its constraints of fuel availability, connection to the network, and political objection to the site itself (eg being too close to a community etc.). Hydro plants have considerably less choice of site as they ought to be built near a water source.
The major hydrological risk is the water flow rate. It is exceptionally difficult to make accurate flow rate predictions particularly given the traditional paucity of historical flow rate information.
Rainfall is unpredictable and it is possible that climate change may make forecasting more difficult. The project company (together with its sponsors and particularly its lenders) will find it difficult to assume the hydrological risks given the potential impact on revenue. If there is insufficient flow, the result is that the plant will generate less power. The use of a take or pay clause effectively moves the hydrological risk to the offtaker. Where there is a government guarantee of offtaker performance, the risk is effectively taken by the government.
What is hydrological risk in relation to electricity generation?
A hydrological risk here refers to the risks of having periods of low rainfall which will reduce power generation and thus revenues. There is some debate as to whether there is an incentive for the flow rate study to exaggerate the results. Higher flow rates mean that output is larger and the economics of the dam should improve. It is also apparent that some dams have been unable to achieve their original projected output (in particular the Akosombo dam in Ghana, and the Aswan High Dam in Egypt) and that in the light of that failure it can be argued that the flow rate predictions were optimistic (although in the case of the Aswan there have also been unforeseen problems with silting).
What is silting?
Silting is a sedimentary material consisting of very fine particles with sizes ranging between clay and sand. Silting in reference to a dam means to fill, cover or obstruct the dam with silt thereby reducing the flow of water. It is the deposition of water-borne sediments in stream channels, lakes, reservoirs, or on floodplains, usually resulting from a decrease in the velocity of the water’. The key point in hydro projects is that there is an expectation that government will assume the hydrological risks. That in turn assumes that there will be a power purchase agreement, in the absence of a PPA, the project would by definition take the risk. A second category of risk which the project company (together with the sponsor and the lender) will find difficult to assume is the risk of insufficient demand on the network.
Most new plants are built to service growing demand, and the rest as replacement for obsolete plant. If that expected demand does not materialize, either because the estimate was too optimistic or because economic conditions turn out to be different from the assumptions in the estimate, the power sponsor will find it difficult to assume the risk in the absence of a transparent and non-discriminatory wholesale market. The policy decision usually permits a degree of interference in the market (taking various types of plant out of the market) or a lack of transparency in the operation of the market. In many countries the policy decision is to have a limited degree of competition, or indeed to restrict liberalization to competition to enter the generation market.
How can the PPA deal with the risk of insufficient demand?
The PPA deals with the risk of insufficient demand by moving it to the off-taker by means of the take or pay clause. It is worth remembering that the take or pay element need not be 100%, but where there is no market it would be surprising if the sponsor (and the lender) were prepared to build the plant in the absence of contractually guaranteed full off-take. No market means there can be no other off-taker. Power from new plant may be more expensive than old (depreciated) plant. In the event that a new plant is built but the power is not needed, the take or pay clause means that the off-taker must still pay for the power. That has unfortunate consequences for the consumer price of power, or for the size of the subsidy required by the off-taker. It also means that if demand has not materialized due to a poor economic climate, the price of power (or the taxpayer subsidy) rises at a time when a stimulus of low electricity prices would be most welcome.
Insufficient demand and the take or pay clause
The effect of take or pay clauses means that there is a strong incentive for the government to get the demand estimate right. Any incentive for government to deliberately underestimate demand and thereby reducing if not eliminating any future take or pay problem is probably outweighed by the accompanying risk of failing to build sufficient capacity to keep the lights on. Signing too many take or pay contracts can get expensive for government (where there is a government guarantee). This may lead to a temptation to refuse to sign any take or pay contract. The problem with that approach is that it fails to recognize why the take or pay clause exists.
The existence of the clause is recognition that there is no alternative offtake route. It implies that the best person to bear the risk of insufficient demand is the government rather than the individual generator. In the absence of a transparent and non-discriminatory wholesale market, the project company really has no option but to seek a take or pay clause, it serves as a guarantee to the project company. There is a question linked to this, namely how to recognize when there are too many take or pay contracts. The answer will be country specific, and the problem is that it is dependent on the reliability of the future demand forecast amongst other factors. The answer also clearly changes over time. There is no doubt that a reliable network needs more generation capacity than is being used and that investment confidence is higher where there is a PPA than where there is not. But there is also no doubt that unused power plants are expensive and a take or pay clause means that the expense falls on the offtaker.
The effect of take or pay may be that the market operator chooses to run the expensive plant and shut down the older cheaper (state run) plants to minimize losses. This has a tendency to cause extreme resentment amongst the existing generators, who can legitimately point out that in a competitive (spot) market the cheapest source should normally be switched on first. The response to that is equally obvious that the PPA means that the project company is NOT competing in a spot market. The generation market is segmented by the PPA’s take or pay clause. In the case of a country where the offtaker is a vertically integrated utility, choosing new plant (PPAs) over old plants means that the offtaker is switching off its own plants. This is so because the operation of the take or pay clause will force purchase at PPA price. The offtaker will not however switch off its plants where its own interests can override contract law, which is a possibility where the rule of law is not rigorously applied.
Exchange rate risks (currency risks)
Exchange rate risks are one of the most awkward issues in power projects and an issue which is impossible to avoid. The issue arises because the costs of most projects are in one or more currency whilst the revenues (the electricity price) are denominated in local currency. Exchange rate movements therefore have a powerful impact on the project. There is an additional complexity in that the issue need not arise immediately, and indeed can appear at any time over the life of the PPA. Exchange rate movements are not directly linked to PPAs; this is an external factor which the PPA must take into consideration. The goal is simple. The PPA aims to ensure that the generator receives the same value of the consideration irrespective of movements in the exchange rate.
The PPA can achieve this either by compelling the offtaker to pay in hard currency, or alternatively to pay a local currency amount equivalent to a given hard currency value. The effect is that the offtaker bears the exchange rate risk and the project company is insulated from the risk. From the perspective of the offtaker, there is a second problem. If the local currency value deteriorates, the price of power under the PPA rises. In turn, that makes alternative local sources look more attractive, and puts pressure on the operation of the contract despite the presence of the take or pay clause in the contract.
Currency risk is the risk that returns on investment will be affected due to exchange control restrictions or change in the exchange rate of one currency (usually a local currency) against another (foreign currency) and usually affects investment across national boundaries. Currency risk with respect to an international infrastructure financing has three components: the risk arising from incurring liabilities to fund a project denominated in a currency (typically us dollars) different from the currency of project revenues and the risk of inconvertibility of, or inability to repatriate, currency arising from exchange controls or other currency restrictions. (Forrester J & Mayer Brown and Platt, Debt Finance for Infrastructure Projects available at: http://www.securitization.net/knowledge/transactions/fi n_debt.asp)
Any devaluation of the local currency in which project revenue is denominated will require a corresponding increase in the project’s cash flow to assure that economic returns are maintained, and to service the project’s debt (especially if denominated in another currency). Any significant devaluation of a local currency, without offsetting factors, will directly impair the project’s likelihood of repaying its lenders and other investors. (Forrester, J and Mayer Brown & Patt- IBID).
There is no better example of the problems which currency risk causes than Indonesia. It was however a special case. There was no formal government guarantee of offtaker performance, as at the time it was not thought necessary given the Suharto regime’s track record of unflinching support for foreign investors. The PPAs were expressed in local currency with a peg to the US dollar. The fall in the value of the local currency was however so great that the offtaker simply could not cope, in effect the local currency became worthless. In these circumstances the offtaker simply cannot perform the obligations required under the PPA. One answer would be to limit or cap the offtaker’s exchange rate risk in terms of the PPA, moving the rest to the project company. That would be unacceptable to lenders given that the project company has no way to influence, let alone control, the country’s exchange rate.
Where the exchange rate deteriorates to the point where the offtaker simply cannot cope, the answer is usually to rely on a government guarantee. There is however a remaining possibility that the government is similarly unable to afford the guarantee. Given that the project company cannot influence the exchange rate, it is usual to see that the (state owned) offtaker assumes the exchange rate risk in its entirety. The argument then turns to how the offtaker controls that risk, which is normally through moving it to government by means of a guarantee. After the Indonesia crisis, it is abundantly clear that track record is no longer enough to satisfy investors and lenders.
In the case where the offtaker is not a state entity, it is in no better position to assume exchange rate risks than the generator. In that case the division of risk must be negotiated. This is not a problem which arises frequently given that countries with independent suppliers usually use hard currency (a happy accident, the two are not directly related) and the exchange rate risk is lower.
Negotiating the terms in a Power Purchase Agreement and conditions of a generation project is fraught with difficulty. The fundamental problem is the information asymmetry; the only party that knows the true costs is the generator. The government cannot know if it is getting the best possible deal or not. That has not stopped numerous power projects throughout the world from being put together on a negotiated basis rather than a tendered basis. The argument is that hydro projects are unsuited to competitive tender, on the basis that the project is location specific. Whilst it is undoubtedly true that hydro projects are location specific, that fact does not mean that the project cannot be competitively tendered. In a tender for a coal fired plant for instance, the tender documents will typically specify the site (location), size (capacity), fuel (perhaps even the source of coal) everything except the price of power. There is no conceptual reason why dams are any different.
The sponsor will certainly ask building contractors to tender to carry out the construction. It is strange that the World Bank did not insist on competitive tender of hydro projects in the same way as it has tried to insist on competitive tender of thermal plants. Negotiation of project terms also leave room for allegations of corruption. Tendering is not a perfect answer to the problems of corruption, as the point at which the corrupt decision is applied may simply move. But tendering which guarantees a measure of transparency makes corruption more difficult by making it more straightforward to detect.