Experience from private hydro development13 April 1998
Øyvind Ulfsby* questions some of the established truths of public and private hydro power in developing countries, using the Khimti plant in Nepal as an example
A private hydro power developer wants predictability, and will ‘cherry-pick’ projects that offer high head, run-of-river, stable discharge, peaking reservoir possibility and moderate tunnel lengths. He will avoid plants on large rivers that require big dams and the inherent flooding and relocation, as well as multipurpose projects with possible conflicts between users. Large projects are also the most vulnerable to environmental risk, which is an uncontrollable factor as NGOs and governments may promote new or retroactive legislation that has serious economic consequences.
Private developers also look for predictable capital costs, but in a hydro project there is a high proportion of civil work (typically 40-50%), which may be difficult to finance locally. The role of multilateral and bilateral lenders is therefore particularly important. As the proportion of the total tariff earmarked for debt service will be high, lenders will be very cautious about the financial risk of the off-taker.
Hydro power plants which are successfully project-financed will typically be 50-500MW, limited at the lower end by project finance development costs, which for small plants are very high, and at the upper end by location and financing.
As the need for peaking capacity and load following is growing faster than simple electricity consumption, foreign investment often gravitates to this type of project.
Within the project organisation the roles of the decision-making groups are fairly well defined, but may change in future as private industry becomes more experienced in hydro development.
Sponsors should be selected based on their experience in hydro power and project financing, as well as their financial strength and long term commitment. Competitive bidding for the selection of sponsors has been tried and, while it can work for thermal power, for private hydro power plants it is in my view a dead end, with no success demonstrated anywhere.
There should not be more than two or three sponsors, of which one must be a strong local sponsor and one the lead sponsor. Shareholding between sponsors should also be reasonably balanced — suppliers with, for example, 5% of equity can find it difficult separating their roles as suppliers and investors.
Debt financing is provided by:
multilateral development banks; bilateral agencies; commercial banks; export credits (including guarantee agencies); local banks; and occasionally institutions such as pension funds. Only 30-40 banks have a tradition of project financing and have the sophistication to analyse such complex projects.
One of the challenges of developing a BOOT project is to provide adequate security to the lenders. While contributing around 70% of the financing, lenders accept very limited completion risk. The lenders have the final say in all questions regarding project structure and risk allocation, so the quality of the lenders is of great importance.
Dependence on multilateral banks is greater in hydro power than in thermal power projects, because of the much higher content of civil work, which cannot be financed through export credits. They also offer protection due to the cross-default effect. The importance of the multilateral lenders could be still more decisive if their catalytic role was given greater weight, eg by guaranteeing the buyer’s obligations.
At Khimti 1, Statkraft was not able to raise local capital to cover those civil construction costs incurred in Nepalese rupees. For this reason and for cost reasons (higher interest and lower maturity on local loans) there were two multilateral lenders — ADB and IFC. The project benefited from risk protection from these multilaterals, but suffered because the lenders were not able to co-operate through the evaluation process, the preparation process, or the negotiation process. As a consequence the process was time-consuming. It was also expensive, as the borrower had to bear the cost of both sides’ lawyers. A further result was the worst possible combination of loan conditions, since neither lender would be ‘nicer’ than the other. Ideally, one lender should have been appointed as a lead bank to co-ordinate and negotiate.
The banks were unpredictable in the time needed to process the loan applications — at a cost to the borrower of anything up to US$40-50,000 per day. The long processing time may, however, be due to the fact that both the sponsors and the lenders were unfamiliar with project financing of hydro power plants. There is room for improvement in the future.
Similarly, the banks’ loan processing methods, schedule and conditions were obscure. IFC had board approval for Khimti, in principle, two years before financial closing, without telling the applicant it had been submitted, or on what conditions. First repayment date and loan terms were fixed, and a commitment fee incurred, two years ahead of construction start. Conversely, ADB’s board approval came very late and delayed the process.
Lenders’ advisors are typically paid by the sponsors, but selected entirely by the lenders. The terms of reference, the selection process, conditions and the cost should be discussed with those who pay for it. The motivation for both the lenders and the advisors to think and act economically has been low — the original cost for the lenders’ legal advisor grew by 1100%, for example.
These factors notwithstanding, the lenders for Khimti must have credit for extraordinary effort and stamina. Without them there would have been no project.
The host government
Private power will only be successful where it is actively supported by the government and the purchasing utility. Too much bureaucracy can kill a project. Complexity, delays, overlapping jurisdictions and endless requests for more useless information characterises the non-successful developing countries. Setting up a central decision-making group is a big step forwards.
The electricity sector cannot in the long run be subsidised. If the government thinks it can choose whether to raise consumer tariffs or subsidise the electricity sector, it is wrong. A long-term requirement for internationally financiable power projects is that the average tariff covers the cost of electricity and is fixed by the market or by a politically independent tariff committee.
The lack of a well-established legal, institutional and regulatory framework in many developing countries is the most important risk facing IPPs. In our experience projects in many developing countries are not financiable unless the host government accepts foreign law for the main project contracts. The financing documents must be governed by a law known to the lenders and investors and the regime must allow for enforceable contracts.
Government guarantees must include access to foreign currency, taxation and compensation for potential law changes and governmental force majeure. It has also been usual that the government guarantees the payment obligations of the state-owned utility under the power purchase agreement, but this may not be possible in the long term, as the cumulative effect may damage the country's credit rating.
The buyer of the power in developing countries is typically a state owned utility, which may be technically bankrupt, may want to retain its monopoly, and may be ignorant of commercial perspectives.
The safest and fastest way to a bankable project is to have single point responsibility through one EPC turnkey contractor, with a fixed price, fixed date contract with performance bond and heavy liquidated damages for delay and performance deviations. However, this may also be an expensive solution, and it has not been very common for hydro power in developed countries.
International competitive bidding is a principle in the World Bank and in many host countries, but there are nevertheless good reasons for questioning the practice for hydro power projects in developing countries.
The intentions are of course the very best: to minimise the cost of electricity and to have a transparent selection process. However, for hydro IPPs selection criteria cannot be established. Bidding is desirable for commodities where specifications can be precise — such as electrical and mechanical equipment — and where the differences between different suppliers are predictable and quantifiable. But for a hydro power plant most work is underground and detailed specifications do not exist (certainly not at the bidding stage). It is even more dubious when the contract structure is an EPC turnkey, where the contractor has the freedom to design the plant according to functional and quality specifications.
The biggest mistake is to select hydro power sponsors via international competitive bidding. No sponsor or contractor can offer unconditional bids, so all the reservations will make the bids worthless.
The ICB process is time consuming in addition to being illusionary. Competitive tendering is delayed because bids are solicited without considering non-cost criteria. The lowest cost bid is likely to be from a bidder who is not credible on other criteria. The result is administrative paralysis, as decision makers struggle to explain why the low-cost bid must be rejected.
The cost of delay is not given enough attention. Although construction delays are penalised, nobody, not even the host countries, considers the cost of unserved energy while development is delayed. The cost may be twice the production cost per kWh — US$100-150M, if a 1TWh plant is delayed by one year. This should be a joint concern of host countries and MDBs.
Using competitive bidding provides only the appearance of transparency and cost savings. It would be faster to select sponsors on their competence and financial strength and allow them to decide whether to use EPC contracting.
Advisors and consultants
The complexity and cost of financial closing increases with the number of participants. Meetings with 30-40 participants, of which half may be well-paid advisors, are not unusual in this business. To make sure all parties aim to reach financial closing quickly and cheaply advisors should be paid on a success fee basis.
The role of the independent engineer (IE) should be reviewed. The IE may see 20 years of profit in the project’s development, construction and operation phases. Having appointed an engineer, the lenders cannot easily discard his opinion, and the IE becomes the real decision maker on technical and environmental issues, and to some degree on economical issues. However, the IE has no real economic responsibility and no incentive to save time and cost.
Some hydro power engineering companies do not have recent practical experience, and may not be familiar with the state-of-the-art, so new cost-effective solutions are seldom recommended. This drives up costs and reduces profitability.
If an EPC turnkey contractor builds a plant based on functional and quality specifications, design and implementation is his responsibility, demonstrated through heavy liquidated damages for performance as well as delay. Lenders and sponsors should therefore concentrate more on specifying function, testing and long-time performance risk and guarantees.
How functional specifications are met through design is the contractors responsibility. By specifying the plant in too much detail, design responsibility may lose focus.
The use of an IE for simple and relatively non-controversial projects should be limited in scope, cost and time. Terms of reference should be very precise and should be worked out together with the sponsors. After all, sponsors and lenders do have the same interest in looking after their investment.
It may be worthwhile to revisit the accepted — very time consuming and costly — development process inherited from the public sector.
An IPP must be evaluated from a technical, environmental and economic point of view. The customer for traditional feasibility studies is the host country, not a potential private developer, and this is often reflected in a lack of financial and risk analysis. The most important part of a feasibility study is the collection of raw data. Most of the completion and operating risk is related to the reliability of these data. Conversely, the analysis and design part of the study often takes too much time, costs too much, and is not based on the appropriate criteria. The EPC-contractor will take on responsibility for the design of the plant. Why then present him with a design made for others, which he will have to redo? Quality, realism and speed should increase, while unnecessary reviews should decrease, if the sponsors and contractors are involved at the design stage. The host country should emphasise collection and reliability of raw data. With the hydrological and environmental information and a pre-feasibility design and cost estimate, a qualified sponsor will be able to assess the project and approach potential EPC contractors.
Statkraft Anlegg has for example delivered a firm fixed-date, ‘not to exceed EPC price’ proposal for a hydro power station based on only a pre-feasibility study, and functional and quality specifications.
Risk management is always important, but never more so than in project-financed deals, where the lenders’ only security is the cash flow from the project and the assets of the project itself. The consequences of any cost overrun, delay and performance deviation must be allocated between the parties (sponsors, buyer, host government, lenders and contractors) before funds are released and construction can start.
A hydro power developer faces unique risks. First, each scheme is unique; the dam, the penstocks, the power house and the generating plant are all designed specifically for that project. Secondly, the geology has been evaluated from limited site investigations and cannot be fully revealed until after construction commences, and the proportion of underground work is very high. Thirdly, the flow is not likely to be known to an accuracy of more than 10-20%.
Cost overruns, delays and poor performance will normally be the sponsor’s risk, managed through fixed-price construction contracts with obligations for liquidated damages, physical contingencies in the construction budget, contingent equity commitments from the sponsors themselves, and advance loss of profit insurance.
At Khimti 1, liquidated damages for delay were capped at 25% of the contract price under the civil construction and engineering contract and 10% of the contract price under the equipment supply and installation contract. Furthermore, the sponsors provided a guarantee up to US$10M in contingent equity, and the project company obtained an advance-loss-of-profit insurance for one year of delays for insurable risks. The civil contractor (here Statkraft Anlegg, a construction company subsidiary of Statkraft), is thus the most important completion-risk taker who carries the first part of potential cost overruns and delays up to the negotiated cap. Thereafter the risk passes to the project company. Lenders are not assuming any completion risk in the project.
As for all run-of-the-river based power projects, electricity output depends on the river flow. The risk is fortunately limited to flows during the dry season (normally 6-7 months) — in the wet season, the water flow will far exceed the power plant capacity. As the hydrological records have been taken by the host country and hydrology is a factor that nobody can control, it would be reasonable that this risk be absorbed by the host country. However, this is an important topic for negotiations, and there are different solutions. In India the buyer takes the hydrological risk for the first seven years of operation, while in Nepal Statkraft took the hydrological risk against a buyer’s obligation to pay for all energy in excess of contract energy available in the dry season.
Meeting the buyer’s obligations
The financial health of the buyer, and the security arrangements for the buyer’s obligations, are the crucial questions in any project financed deal. Typically, the buyer is not creditworthy and different methods for securing the revenue stream are necessary.
At Khimti the buyer's obligations under the PPA are supported by the collateral arrangements under the PPA itself, and by government guarantee. An uncon-ditional Letter of Credit (LC) will be maintained by the buyer in favour of the project company for three months’ projected pay-ments of the demand charge. The LC will be replenished by the buyer within 30 days following any draw made on it by the project company . The government guarantees that the buyer shall perform all of its obligations under the PPA in a timely manner. In the event the buyer fails to pay its obligations under the PPA, the government makes payment directly to the project company. In addition, the government will assure availability of all necessary foreign currency and repatriation rights of all of the buyer’s payments.
This example is unlikely to continue to be typical. The most important catalytic effect the World Bank could have for IPPs would be to increase its efforts to alleviate the credit risk of the buyer.
Political risks range from delays in obtaining permits and licences to deliberate expropriation by a host government. Gaps and residual risks not covered by the host government must be managed by the sponsors.
In Nepal, political risk is managed by:
•Local investors’ participation in the project.
•Involvement of multilateral financial institutions.
•Political risk insurance from multilateral institutions.
Political risks are, however, not easy to insure against. There will always be uncertainty regarding government support; inefficient legislative and regulatory systems; and general economic and political insecurity. The political risks may in some countries be potential landmines, and will remain so in the transition from public to private ownership. The World Bank and the regional development banks can play a greater role in mitigating such risks, and thus emphasise their catalytic role as development banks. Today the commercial role seems to be given most weight.