Norton Rose Fulbright has been dedicated to the development of energy markets across Africa for more than 30 years. With a reputation for completing complicated energy projects across Africa within demanding timeframes, we can put at your disposal an integrated and experienced team of energy lawyers from across our 850-strong network.
We have advised on all aspects of the acquisition, development, construction and financing of power plants and transmission projects. Our experience includes advising sponsors, lenders, developers and governments in Africa.
We have particular experience of advising on captive power projects and industrial supply arrangements throughout Africa. We set out below a selection, which includes advising
- A mining company in relation to its power purchase arrangements from an on-site captive power plant in Ghana
- A steel producer in relation to the development, construction and financing of a captive power plant in West Africa
- A bidder in relation to its bid to develop a hybrid solar and HFO captive power plant in Tanzania
- Lenders in relation to the financing of a bioethanol plant and associated captive power plant in Sierra Leone
- Lenders in relation to the financing of a captive power plant supplying a steel mill in Nigeria
- An industrial estate owner in relation to the development, construction and financing of an embedded generation power plant in Nigeria
- An offshore logistics base owner in relation to the development, construction and financing of an embedded power plant in Nigeria
- An international developer in relation to the development, construction and financing of a captive power plant supplying a beverage production plant
- An international developer in relation to the development and construction of a captive power plant supplying a brewery in Kenya
- In relation to power supply arrangements for two mines in the Democratic Repubic of Congo
A captive power plant is a plant that supplies power wholly or primarily to one industrial customer (or a limited number of industrial customers), rather than to a utility. In many countries the term “captive” is used interchangeably with “embedded” generation. However, this is not always the case. In Nigeria, for example, captive power plants are those where output is consumed by the owner of the power plant whereas an embedded generation plant is one where the owner sells the power generated to third party customers. For the purposes of this paper, we do not distinguish between captive power plants and embedded generation and instead, focus on projects where a private generator sells power directly to one or more industrial customers. Even in the Nigerian context, the discussion below will be of relevance to embedded generation projects. Why
There are a number of reasons why a captive power plant may be developed in preference to obtaining power supplies by connecting to a grid.
Many industrial users, such as mines, steel mills, refineries, fertiliser plants, etc. have significant power requirements and in developing economies, the national utilities often do not have the capacity to supply power on a stable and reliable basis. Indeed, in countries facing power shortages, it is often the more intensive industrial users that are not supplied during periods of load-shedding.
Commercial reasons for developing a captive power plant generally centre on a need for a high level of control over the development, construction and operation of the power plant, where these needs cannot be met reliably from the grid, either because new generation capacity is needed anyway or because of doubts about the physical condition or management of the existing system. A captive power plant provides these customers with security of power supply, coupled with greater control of increases and decreases in generating capacity, to align with their production cycles.
Technical reasons for developing a captive power plant can also include the offtaker’s need for other products that can be provided by a cogeneration plant, and a remote site location where a grid connection is not feasible.
For example, cogeneration plants can provide heat to the offtaker in the form of steam, which is required by petrochemical refineries and paper mills. Plants can also be configured to provide other products, such as water desalination to a dedicated offtaker. Another example is a captive power project supplying a bottling plant, where the carbon dioxide produced by the power plant was purified to food grade and then used to produce sparkling drinks that were then bottled by the plant.
A further point to note is that captive power plants tend to be of a much smaller size than power plants that are constructed to sell energy to a utility, the reason being that the power needs of one (or a few) industrial users are limited, but this is not a hard and fast rule. We have seen captive power plants of 1MW and also 50MW.
In most respects, a captive power plant project will be structured like any other independent power project. It will require a long-term power purchase agreement (PPA) to support the developer’s equity investment and debt financing. However, given that the essence of a captive power project is the sale of power from a private generator to a private offtaker, the parties generally have significant discretion as to how the project and the key agreements will be structured, although we make some general points below.
The offtake agreement will generally cover any other products to be supplied by the plant, such as steam or water, in a single document. The agreement may be a tolling agreement or the generator may source fuel from a third party supplier. For the most part, it will be on standard terms for a PPA for an independent power project.
If there is a small group of offtakers rather than a single offtaker, there would be separate PPAs for each. A project coordination agreement between the generator and all of the offtakers may be required to deal with common issues, such as the allocation of curtailed output. Consideration will also need to be given as to whether the terms of the PPA (including the tariff) will differ between small offtakers and large offtakers.
The creditworthiness of the offtaker must be considered. The offtaker is often a special purpose company raising project finance for its own project. Its ability to take electrical output and pay the tariff will be contingent upon the success of its own project. Guarantees from its ultimate sponsor may be insufficient from a credit perspective and the use of standby letters of credit or bank guarantees to support the offtaker’s tariff and liquidated damages obligations to the generator is quite common.
There may also be a separate concession agreement between the generator and the host government, although this is by no means a pre-requisite for most captive power projects. As well as setting out any tax concessions and foreign investment protections, this can be a useful means for the generator to transfer political force majeure risks to the government, as it is unlikely to be realistic to transfer such risks to a private offtaker or to the generator’s subcontractors.
The tariff for a captive power plant must be competitive from the offtaker’s point of view, although the security of a long-term dedicated supply from a new plant, with all of its ancillary infrastructure, will necessarily come at some premium. For offtakers with a very heavy power demand, such as aluminium smelters, the electricity tariff will be the major determinant of the viability of the offtaker’s wider project.
As with all independent power projects, the tariff must support the cover ratios required by the generator’s financing and an acceptable equity return. It must also incentivise the availability of the plant. This is usually achieved by penalising poor availability with tariff deductions. However, PPAs for captive power plants sometimes have output-based rather than availability-based tariffs, so the penalty may take the form of liquidated damages payable by the generator to the offtaker, sized to cover either the offtaker’s cost of alternative power supplies or some element of the offtaker’s lost production.
The tariff may be structured as a traditional availability charge and energy charge, whereby the generator’s fixed costs are covered in the capacity charge and variable costs will be covered in the energy charge. Fixed costs will include debt service, equity return and fixed operation costs, amongst other things. Variable costs will include fuel costs, variable operating costs and any other costs that vary depending upon the amount of power generated. The availability charge would be payable for generating capacity made available, regardless of whether the power consumer actually takes power – in other words a “takeor- pay” arrangement. This type of tariff will typically be escalated on an annual basis by reference to a price index.
Another form of tariff is a fixed price for each kWh delivered to the customer. There would typically be a deemed energy payment where the customer fails to take delivery and there will be deductions from the monthly tariff charge for curtailment of generation. In order for such a tariff to be viable for the generator, it needs to have a secure fuel supply and the price per kWh needs to have sufficient “cushion” within it to deal with increases in fuel costs. There could be a renegotiation clause if the price of fuel increases above a certain level.
A captive power plant may be connected only to the offtaker’s facilities via a private network or it may be connected to an interconnected transmission or distribution system and capable of parallel operation with that grid. This has a number of important commercial implications.
If the plant is not connected to a transmission or distribution system
- There may be insufficient load to conduct handover tests on the captive power plant if construction of the offtaker’s facilities are in delay and the PPA must give the generator adequate relief in this respect.
- The offtaker does not have the option to obtain power from the grid when the captive plant is unavailable, which means the generator will have to assume responsibility for a very high level of reliability under the PPA.
- There will be limited scope for the generator to schedule outages of the captive plant and these will need to be coordinated with outages of the offtaker’s facilities.
- If the offtaker defaults, the generator will not have the option to sell power to third parties, with the risk that it becomes a stranded asset.
- The captive plant will need to be configured for island operation, with black-start capability.
If the captive plant is to be connected to the utility’s transmission or distribution system, the generator will need to enter into a grid connection agreement and may also need to enter into a wheeling agreement whereby the utility will charge the generator for the use of its grid, in delivering power to its customer.
The high level of reliability demanded of a captive plant may require some redundancy (an additional generating unit) to be incorporated in the captive plant. This has a high capital cost and the right balance must be struck between cost and reliability.
If, on the other hand, the plant (or the offtaker’s complex) is connected to a transmission system, this will mitigate all of the above concerns, although this will inevitably trigger a discussion of who takes the risk of transmission. Is the responsibility of the generator to deliver to the grid or to deliver to the connection point where the offtaker is connected to the grid? A generator will generally resist any attempt that it should assume the risk of transmission losses.
In countries with a single buyer electricity market, the owner of a captive power plant may require a generation license to sell power to a private offtaker.
Where the offtaker is developing a wider project at the same location, it is likely that the offtaker’s environmental and social impact assessment and environmental permitting process will include the captive power plant. This can be problematic where the offtaker’s assessment is already at an advanced stage, before the generator and power plant technical solution have been selected. The offtaker and the generator will need to cooperate closely in this regard.
Where the generator is supplying power to a number of customers, it may be that the generator also constructs a localised distribution grid. In such circumstances, the generator may also need to obtain a distribution license.
Where the captive plant is part of a wider industrial complex, the plant site is often made available by the offtaker to the generator under a sub-lease. Consideration must be given to what happens to the captive plant
- If the offtaker’s head lease is terminated or
- If the PPA is terminated
The PPA and the lease must be coordinated in this regard.
If the wider complex and the captive plant share certain services (such as security, fencing, fire-fighting or drainage), this will need to be documented.
Insurances are occasionally placed on a site-wide basis and this can be a very complex issue.
The treatment of force majeure is an issue that will need to be addressed on a project-byproject basis, particularly the treatment of political risk. Whilst we would generally expect each party to bear the consequences of “natural” force majeure (such as fires, floods, etc.) affecting its ability to generate or to accept power, the risk allocation for political events (such as government actions, changes in law, etc.) will need to be negotiated and agreed on a caseby- case basis.
What happens if the PPA is terminated, whether for offtaker default, generator default or for other reasons? A termination payment from the offtaker to the generator would normally be required to support non-recourse financing for an independent power project, so that the lenders will be made whole in such circumstances. The termination payment would normally need to cover the generator’s senior debt in all circumstances, with the extent of any equity compensation depending on the reason for termination.
It can be difficult to achieve a robust termination regime for a captive power project, in that the offtaker may not be able to meet a termination payment where it is in default and is unlikely to be supported by a government guarantee. Rather than one perfect solution, the generator and its lenders will probably need to rely on a combination of mitigants, which might include
- Security for the offtaker’s termination payment, such as a standby letter of credit, albeit subject to a cap.
- Suspension of the obligation to deliver power where the offtaker is in payment default, in order to contain fuel costs.
- Retention of the power plant by the generator with the possibility of selling output to third parties where there is a grid connection.
- A termination payment equivalent to a number of years of revenue payable under the PPA, rather than for the full amount of the debt. This payment could allow the generator to service its debt for a period when it seeks to sell to alternative customers, including through the construction of transmission lines to connect to the grid.
- The right for the generator to mothball or scrap the power plant, to strengthen its bargaining position against the offtaker’s lenders.
- Remedies against the host government under a concession agreement and political risk insurance cover in relation to termination arising from political risks.
In situations where there is a small group of offtakers, there is a possibility that one will default, resulting in the termination of its PPA, while sales to the non-defaulting offtakers continue. There are at least two possibilities to deal with this. First, the remaining offtakers could increase the volume of electricity that they take. Alternatively, the defaulting offtaker could be required to make a termination payment that is sufficient for the generator to prepay part of it debt, so that the project can continue on the basis of the existing PPAs with the continuing offtakers.
Much will depend on whether the captive plant can sell power to alternative customers. If so, it would be usual to see a termination payment from the consumer which covers a number of years of projected payments under the PPA. However, in circumstances where the captive plant would effectively be a stranded asset following termination of the PPA, we would expect the termination payment from the customer to constitute a buy-out of the plant from the generator, covering the generator’s outstanding debt, equity and equity return.
All of the key issues one would expect to be relevant to the financing of a conventional independent power project selling to a utility will also be relevant to the financing of captive plants. The viability of the project will depend on the credit strength of the offtaker. However, given that the offtaker will typically be private, rather than state-owned, the structure of the financing will need to take into consideration the fact that World Bank partial risk guarantees and political risk insurance may not be available to back the financing. In such circumstances, there is an increased likelihood of lenders requiring some form of recourse to the generator’s shareholders. It is therefore more difficult to structure captive project financings on a non-recourse basis and they are sometimes funded on an equity or mezzanine debt basis, with the intention of re-financing the project when it is operational. The fact that many captive power plants are small can also give rise to a structure whereby a generator funds a number of such plants, each selling to different offtakers, on an equity or mezzanine debt basis, and then refinances the portfolio with a longer term limited or nonrecourse financing.