South Africa is facing a growing energy crisis. In the absence of significant proven gas reserves, it will need to import LNG to support the development of power generation capacity. With the expected growth in worldwide LNG supply over the next few years, now is the optimum time for South Africa to have launched its gas-to-power programme.
The South African Department of Energy recently issued a Request for Information (RFI) regarding possible developments in a proposed Gas-to-Power programme. This RFI has its genesis in the Integrated Resources Plan 2010, which determines as national policy a requirement to procure new electrical power generating capacity from gas (amongst other energy sources) in order to address the mounting power crisis in South Africa and to fulfil a long-held desire to build a gas based economy. Although coal will continue to play a major role in meeting South African energy demand, environmental problems and the resulting costs associated with its use have been of concern for some time.
No such gas based economy presently exists in South Africa, and the country is caught in an impasse between developing domestic demand for gas (which depends on gas being readily available) and developing gas distribution infrastructure (which depends on the existence of local demand). The Gas-to-Power programme is seen as a way to break this impasse.
The RFI speaks boldly to respondents providing information on possible Gas-to-Power solutions. These solutions could take two forms. First, there could be a ‘cradle-to-grave’ sourcing of gas, comprising development of the gas import infrastructure and delivery of the gas to, and processing of the gas in, that infrastructure, then utilisation of that gas for power generation and sale of power to Eskom under a long term power purchase agreement (PPA). Alternatively, there could be solutions that speak to discrete elements of this value chain.
In this briefing, we set out certain issues in a South African context that may need to be addressed in order to facilitate the success of this proposed Gas-to-Power programme, as well as more general issues associated with Gas-to-Power projects that will need to be considered by interested developers and lenders. It is hoped that the Gas-to-Power programme will follow in the footsteps of South Africa’s highly successful renewable energy programme which has resulted in nearly 4000MW of renewable power capacity in the space of only a few years.
Demand for gas as feedstock for power generation is projected to rise considerably with a number of new power stations being planned in Coega and Richards Bay, for example. There are also various ageing diesel and coal-fired power stations that could be converted to gas.
In the longer term this increase in demand is expected to justify the introduction to South Africa of gas from northern Mozambique through the 2,600km main pipeline that is under development by SacOil. The RFI also provides for the possible use of indigenous gas. However, the RFI recognises that neither will be available in the short-to-medium term and that a successful Gas-to-Power programme will for now depend on the import of liquefied natural gas (LNG). Although South Africa of course has existing gas infrastructure, it has no LNG importation capability so this will need to be developed. The import of LNG will not exclude other feedstock options in the longer term but, because LNG regasification and import infrastructure can be built on a modular basis and because of its lower risk and smaller project cost compared to pipeline gas, it is presently seen as being the more feasible option.
A notable feature of the RFI is that Transnet, the state-owned company that owns and operates all ports in South Africa, is not mentioned. It is known that Transnet understands the imperatives of developing LNG import infrastructure and that such development has been on various port plans for several years. What is not known is Transnet’s intentions in these respects, in particular whether it intends to build, own and operate such infrastructure itself (e.g. an onshore LNG regasification terminal or an LNG floating, storage and regasification unit (FSRU) and, in either case, related port infrastructure), or whether it will facilitate third party development across its existing facilities. If the former, this may present a challenge to ‘cradle-to-grave’ propositions, as there will be a key part of the chain over which developers will have no control. This could create risk in terms of seamless integration with other critical paths such as LNG import, downstream transportation of regasified LNG and commercial operation of the power plant.
It is not unusual for state companies to develop their own LNG import infrastructure, often in conjunction with international developers (for example, as we have seen in Spain and as is currently the ongoing case in Bahrain). If Transnet were to follow this approach, in order to ensure the success of an integrated Gas-to-Power solution in South Africa, it would need to commit to providing the regasification facility on time and within budget and to making sufficient regasification capacity available to LNG importers/Gas-to-Power developers on arm’s length contractual terms.
From a financing perspective, if the Gas-to-Power chain were financed as a single package, there should ideally be a neutral allocation of risk between LNG/gas supply and the power plant. Any involvement by Transnet, or any other third party, in the development of any of these elements of the Gas-to-Power chain could present challenges to that risk allocation, since it may be reluctant to assume certain of the risks associated with the importation of LNG, the operation of a regasification facility and the delivery of regasified LNG as feedstock for power generation. For example, Transnet may have difficulty in accepting significant liability in the event that the regasification facility is unable to receive and regasify LNG imports, whereas any failure to do so will clearly have significant commercial ramifications for the LNG buyer (in terms of take or pay liability), for the downstream gas supply (in terms of send or pay liability) and for the power plant itself (in terms of potential deemed unavailability and therefore potential de-rating under its PPA with Eskom). An independent developer that is in control of all elements of the Gas-to-Power chain may be better placed to manage and ultimately mitigate these risks. That said, an advantage of interposing a state company such as Transnet in the supply chain is that it is long-established, has extensive local market knowledge, and its counterparties may benefit from some government support in respect of the risks it is willing to undertake.
Transnet has a good understanding of the Gas-to-Power programme and the need to import LNG, and is expected to be co-operative in facilitating the development of gas as a strategic fuel supply option. However, clarity is still required as to Transnet’s intended role in the development of the required LNG import infrastructure. It is likely that Transnet’s final plans in this regard will depend on the procurement framework to be developed by the Department of Energy following its review of the responses to the RFI.
Aside from the nature of Transnet’s involvement, securing a reliable long-term LNG supply is likely to be the key issue to be resolved for projects to be successful under the proposed programme, particularly given the current scarcity of available indigenous gas in sub-Saharan Africa. Lenders will, of course, expect comfort that a project which it proposes to support under the proposed programme has a sufficient volume of committed gas supply that will at all times meet at least the base case dispatch requirements under the PPA. Given that there is unlikely to be any back-up gas supply available to the power plant, it is likely that a dual fuel option (i.e. a power plant that can run on either gas or diesel or other heavy fuel oil in the alternative) will be necessary, although this of course presents its own risks that would need to be allocated through the supply chain such as increased commissioning, operation and/or maintenance costs and accelerated degradation for the power plant.
A clear advantage of involving international LNG developers or aggregators in a Gas-to-Power project would be that they have access to a portfolio of LNG supplies. A number of large suppliers have already expressed interest in supplying LNG as feedstock for power projects in South Africa. Coupled with the significant volume of US-produced LNG that is shortly to come to market, it is expected that this may all translate into reliable long-term LNG supply for Gas-to-Power projects, without the LNG sale and purchase agreements (SPA) necessarily being dedicated to any one or more specific liquefaction plants.
If the gas project company or one of its affiliates is responsible for sourcing the LNG (under a long term LNG SPA potentially supplemented by spot LNG purchases), it would of course be subject to take-or-pay obligations under its LNG SPA which would need to be adequately passed-through to Eskom under the PPA to the extent that the power plant is dispatched at reduced levels that trigger this obligation. This risk of reduced levels of dispatch could also be mitigated by the gas project company exercising any rights of downward flexibility under its LNG SPA (i.e. a right allowing the purchaser to reduce its annual contracted quantity by a limited percentage).
Although the RFI has indicated that the power solution will provide mid-merit energy, the constraints from the LNG supply side in respect of the need to sign long-term take or pay contracts with minimum fluctuation of LNG volume purchases, will probably dictate that the power project would need to operate and be dispatched at a higher base case load factor. Higher load factors should ensure a lower cost of gas for end users.
To the extent there are concerns about the power plant being unable to take the annual take or pay quantity of LNG as feedstock, the gas project company might even enter into a supplemental gas supply agreement with one or more independent third parties (such as local industrial consumers) whereby each such party may take, on an interruptible basis, surplus LNG/gas. Additional revenues earned under such an agreement could be used to create a contingency against downside scenarios affecting power offtake and would also help mitigate the risk that Eskom may not provide LNG take or pay coverage where the take or pay liability arises due to a forced outage at the power plant (or even at the regasification facility) as opposed to in circumstances of low dispatch. Dispatch risk itself can be further mitigated through PPA revenues being based on available capacity, meaning that the economics of the power project should not be radically affected.
If a third party such as a state entity is responsible for sourcing the LNG, this may help reduce the take or pay risk on the developer, but the developer is likely still to be expected to share or even assume this risk where the take or pay liability arises due to its default (e.g. poor operation of the power plant). In addition, allowing a third party that is not part of the developer group to source the LNG may expose the Gas-to-Power chain to the risk of misalignment or third party default, which could prejudice the integrity of the gas supply stream. If the third party gas procurer is procuring gas for other competing projects as well, for example if the LNG import and regasification project is established with a view to supplying regasified LNG to multiple power plants, the feedstock gas supply could be reduced in favour of a particular project in the event of a failure to supply/force majeure under the LNG SPA or if there is a political requirement to favour one power project over another. In order to protect against this an agreement setting out an appropriate priority of supply arrangement would need to be entered into.
Any LNG supplier will of course require a credit worthy offtaker. If a special purpose company is established as the LNG buyer then it will be required to provide credit enhancement, such as by way of bank letters of credit or parent company guarantees. This issue may not arise though, if a state entity having an appropriate credit position instead assumes the role of LNG buyer.
Closely linked to the question of ‘who’ will undertake the development of the LNG import infrastructure and import the LNG, is the issue of ‘where’ will this be undertaken? The position of the land on which the receiving terminal will be sited is critical as the terminal will ideally need to be built as closely as possible to the regasification facility. There are a number of South African ports where such infrastructure can be developed, however each has its own particular challenges.
For instance, Saldanha Bay is presently undergoing a feasibility study by PetroSA (the South African national oil company) for import of LNG to support its gas-to-liquids facility at Mossel Bay. However, there are some distinct environmental hurdles to be overcome, given the proximity of the port to the West Coast National Park with its pristine Langebaan Lagoon (a Marine Protected Area).
Another concern is space for facilities, both from the marine side (including an FSRU and, if this option is adopted, a jetty, breakwater and pipeline to shore) and the power generating side. Given present port configuration and development, should development be sought at Saldanha Bay, it is likely this would be an offshore FSRU, with the power plant located outside of the port precinct. However, land in and around Saldanha Bay is at a premium and is not readily available. Consequently, state intervention may be required to enable developers to secure appropriate space to develop their power plant and associated infrastructure. Easements or other similar rights of way will also be required in order to run the necessary gas pipelines from the regasification facility to the receiving terminal and then to the power plant itself.
Exacerbating this contest for space is the current development of LPG import infrastructure at Saldanha Bay. The possible dramatic curtailment of petroleum refining capacity in South Africa over the next 2 to 5 years could mean up to a 70 per cent reduction in LPG production and availability. LPG is Transnet’s current gas focus at Saldanha Bay, not LNG, and in the event of the need to import LPG, competition for space and resources at South African ports might favour LPG.
Richards Bay is an ideal location for new gas power plants. Richards Bay is ‘energy-hungry’ due to the nature of the heavy industries in the vicinity. However, Richards Bay is also water-starved, and the significant cooling water requirements required for power generation will most likely have to be supplied by a desalination plant. Richards Bay is an excellent deep-water port that can easily accommodate large bulk carriers, although it will be necessary to identify areas within Richards Bay which are suitable for LNG terminal development. Unlike at Saldanha Bay, it may be possible to construct an onshore LNG regasification terminal at Richards Bay.
Mossel Bay was considered originally for development of LNG import infrastructure, given the direct access to the PetroSA GTL plant. However, this has been shelved during feasibility studies owing to environmental concerns and the hazardous marine environment in which the offshore LNG FSRU would have to be developed and operated.
If an offshore regasification solution is adopted, it will be necessary to consider issues arising from the use of an FSRU. In particular, developers will need to decide whether the FSRU will be chartered under a bare boat charter, a time charter party, or owned and operated outright. If the FSRU will be time chartered, then the developer’s obligations will largely relate to paying hire and issuing directions to the FSRU owner/operator. Chartering will insulate the developer from certain legal risks associated with ownership. However, it will create additional interface risks with the FSRU owner and possibly its lenders. Regardless of whether an offshore FSRU or onshore regasification terminal solution is adopted, if Transnet or another third party were to own and operate the regasification facility, the gas project company would require a tolling agreement with the owner/operator of the regasification facility in order to ensure that a committed supply of regasified LNG is available as and when required as feedstock for the power plant.
A particular challenge under the South African Coal Base Load IPP Procurement Programme has been the question of alignment of adjustments of the energy price under the PPA with the realities of variations in the fuel supply price. At present, the coal IPP programme PPA stipulates that the energy price is indexed annually in line with CPI, with certain further adjustments for certain variable costs. Excluded from the variable cost basket is any variation in the cost of producing coal, and consequent flex in the coal supply price. This is hampering the ability of independent power projects to conclude the long term coal supply contracts that they require to cover the 30 year PPAs with Eskom.
In the international LNG/gas market it is unlikely to be possible to contract for long term supply by fixing a base price and then escalating that solely on a fixed index basis. For example, LNG prices are typically adjusted by reference to various markers including competing fuels (e.g. gas oil, fuel oil) or in the case of US sourced LNG, on the basis of Henry Hub, with only a small part of the escalation being linked to US CPI. Even then, changes in US CPI may of course differ significantly from changes in its South African counterpart. Consequently, the South African Department of Energy will need to address this issue by ensuring that the adjustment of energy prices under the PPA (and therefore a key part of the revenue stream for the Gas-to-Power chain) is sufficiently flexible to allow for price flex under long-term LNG supply arrangements.
In addition to these matters, it will be necessary to consider classic risk allocation issues such as delay in construction of the power plant and associated infrastructure. For example, in the event that liquidated damages are payable by the power project for a delay in achieving the commercial operations date under the PPA, these will need to be flowed back through the project documents so that any element of the Gas-to-Power chain that is not ready on time (for example, the regasification facility) bears that risk. Force majeure and change in law risk will also need to be flowed through the project documents. Ideally, the PPA will not exclude supply-chain force majeure claims and will include LNG-specific force majeure provisions, thereby allowing the power project to claim deemed availability in such circumstances. The management of project risks such as these throughout the Gas-to-Power chain will of course be assisted if, as the RFI contemplates, gas supply and demand can be brought on stream at the same time and spread geographically throughout South Africa.
LNG imported into South Africa is likely to be priced in USD, whereas revenue received under the PPA will, in all probability, be Rand denominated. These revenues will ultimately be utilised by a Gas-to-Power project to make payment for its LNG purchases. It is unlikely that suppliers of LNG will take Rand risk. It may be possible to hedge this commodity price risk for five years, but thereafter it is unlikely that there would be a forward market for a 30 year PPA.
No LNG supplier or investor could, in our view, assume such risk. LNG prices will certainly change over 30 years and it will be necessary to consider means of mitigating risk on the USD portion. One option may be by way of a pass through mechanism incorporated into the tariff, which is an effective USD adjustment.
There is, of course, the possibility (highly unlikely) that revenues under the PPA would have a USD component which would allow for USD tranches of debt and USD lenders, thus diversifying lenders and possibly not only mitigating the risk of the USD denominated LNG purchase but also creating additional liquidity in this market.
There are some regulatory issues that arise under South African legislation, notably in terms of the Petroleum Pipelines Act 2003 and the Petroleum Products Act 1977, and it is possible that legislative amendments or adjustments will be necessary.
For instance, the extent to which third party access is required to be given in the regasification facility and related infrastructure is important, regardless of which party develops and operates these assets, since this will ultimately affect the committed regasification and gas throughput capacity that is available for the Gas-to-Power chain. Under the Petroleum Pipelines Act, third party access is required for uncommitted pipeline or storage capacity, both of which are critical parts of gas import and reticulation infrastructure. However, a Gas-to-Power project cannot be placed in the position where delivery of its fuel is hampered or curtailed by virtue of third party failure at any point in the Gas-to-Power chain.
The Petroleum Products Act has a complex product import control regime which regulates the import of defined petroleum products, including natural gas and petroleum gas. Under the current framework, an importer may not import gas unless it is a licensed manufacturer or is a ‘Historically Disadvantaged South African’ (HDSA) as defined under the legislation. It will be necessary to clarify whether LNG regasification will be considered to be ‘manufacture’ under the Petroleum Products Act. If LNG regasification does constitute manufacture, the prospective importer will need to first apply for and hold a manufacturing licence. Once that licence is issued, the prospective importer will have the right to apply for an import licence. It is probable that an import licence would be issued following the issue of a manufacturing licence. If, however, LNG regasification does not constitute manufacture, then amendments will need to be made to the Petroleum Products Act in order to permit the import of LNG. This may cause some delay and uncertainty amongst potential developers and lenders alike.
Although there will be various challenges in developing a Gas-to-Power project in South Africa, none of these are insurmountable. The development and financing of gas-fired power plants is well understood by developers and lenders, as too is the development and financing of LNG import infrastructure. The key will be to secure long-term LNG supply and to ensure an appropriate risk allocation through these elements of the supply chain ending ultimately with Eskom. International practice and precedent will no doubt play a significant part in developing the contractual and legal structure and, given the increasing demand for power in South Africa and across the African continent generally, once the first such project is under way, others will no doubt follow quickly.