In the face of continuing deficits across the world in power generation capacity, we are seeing increasing consideration of liquefied natural gas (LNG) to power projects as a relatively rapid way of adding
significant capacity to the grid.
This article sets out the background to the rise in the potential opportunities for LNG to power projects and examines some of the specific commercial issues that will shape the development of an acceptable (i.e. bankable) project structure.
There are multiple factors contributing to the serious consideration now being given to LNG to power projects in a variety of locations. One of the main drivers is the falling cost of LNG which allows it to compete with other fuels in many markets. The number of large LNG projects that have recently come on- stream (e.g. in Australia and the US) or are scheduled to do so in the near future, combined with the global gas glut caused by the development of unconventional resources, have driven LNG prices steadily downwards (see figure 1).
There is also the diminishing attractiveness of coal-fired generation, one of the often cheaper alternatives to gas-fired generation, largely due to environmental considerations. Based on the outcome of the recent COP21 conference in Paris and the associated commitment by the global community to reduce carbon emissions, the trend away from coal-fired generation continues.
Figure 1: Asian LNG Spot prices
Environmental considerations weigh particularly heavily on export credit agencies, who have in the past been among the major funders of coal-fired projects in developing markets. In late 2015, those OECD countries party to the official arrangement for export credits agreed new rules that will take effect in January2017 and will limit the ability of those countries to provide finance in support of investments in coal-fired plants. A number of commercial banks have followed suit. This represents a significant policy shift away from support for coal-fired generation, particularly in Asia.
Finally, the technology and commercial viability of floating storage and regasification units (FSRUs) has been proved up in recent years. FSRUs can now present a quicker and potentially more flexible solution to the provision of gas storage and regasification facilities than alternative land-based facilities. Indeed, in the last two years, FSRUs have been brought into operation in countries including Brazil (in two separate locations), Lithuania, Indonesia, Kuwait (replacing a smaller FSRU), Egypt, Jordan and Pakistan.
Many of the opportunities for gas to power are emerging across Africa. This is perhaps no surprise given Africa's deficit in generation capacity and the lack of gas grid infrastructure to support the development of conventional independent power projects (IPPs).
For example, Egypt has recently chartered two FSRUs and is reported to be planning to charter a third.
Elsewhere in North Africa, Morocco has recently launched a request for proposals to appoint advisers in relation to a project for the import of 5 bcm, the development of the associated LNG import and pipeline infrastructure, and two 1,200 MW Combined Cycle Gas Turbine (CCGT) power projects – one at Jorf Lasfar and the other at Dhar Doum.
In sub-Saharan Africa, potential project sponsors are evaluating an LNG to power project in Ghana (notwithstanding the ongoing development of domestic offshore gas reserves). The 1,300 MW Ghana 1000 project led by Endeavor is intending to develop a CCGT in several phases, with the later phases being fuelled by LNG supplied by the Ghana National Petroleum Corporation. South Africa also has ambitious plans to add over 3,000 MW of gas-fired capacity based around LNG fuel supply. More details are expected later this year.
Situated between North Africa and Europe, Malta is in the midst of developing an LNG to power facility. Malta Gas and Power Limited is developing a 200 MW gas- fired power plant at the Delimara power station near Marsaxlokk. The project also involves the development and construction of LNG receiving, storage and regasification facilities for LNG. A floating storage unit will be leased on an 18-year term and the plant will benefit from an 18-year power purchase agreement (PPA) off-take.
Further afield, in Chile, sponsors including EDF and Cheniere are developing an LNG receiving terminal and an associated 640 MW CCGT. LNG will be supplied by Cheniere under a 20-year LNG sale and purchase agreement (SPA), while the project will include a FSRU chartered from Hoegh LNG under a 20- year charter. Likewise, three Colombian power companies have awarded Sociedad Portuaria El Cayao S.A. E.S.P. a contract to develop a floating LNG import facility, which is currently under construction.
Power plants, developed, constructed and financed on an IPP basis, have a long history of being successfully financed. The construction of a single well-understood asset with limited contractual interfaces leading to the delivery of a steady revenue stream, backed by a long-term sovereign or quasi-sovereign credit, has proved one of the more bankable asset classes.
Clearly, one of the challenges for LNG to power projects is the co-development, and potential co-financing, of the LNG and power infrastructure. Not only do LNG to power projects potentially suffer from "project-on-project" risk due to the interdependency of the construction and commissioning of the gas and power infrastructure, but the project(s) are altogether more complex and require a number of additional risks to be considered and allocated, including potentially flowing various risks through a much longer project contract chain.
There is no set single structure for the development and financing of LNG to power projects, and indeed there are a number of factors that sponsors and their funders will need to consider at the outset that may influence the choice of structure, including the following:
- Size: what is the size of the proposed facilities both in capacity and dollar terms? Based on size, would it make sense to undertake a single or two separate (but linked) debt raisings? If two debt raisings are planned, will they be “stapled”, i.e. with the same banks holding the same percentage participations in both loans?
- Identity of sponsors: do the potential sponsors all wish to invest equally in the gas and power generation infrastructure, or would it be beneficial to develop a structure that allows ownership in different proportions or indeed to allow different investors?
- Space: are there space constraints on land for gas storage infrastructure (that would suggest an FSRU-based solution may be more appropriate) or are marine conditions/facilities inappropriate (so as to suggest a land-based approach to gas infrastructure)? An FSRU-based solution will involve the charter of the FSRU, thus adding to the project contract chain as compared to a land-based project.
- Local gas demand: is the gas demand specific to the proposed IPP or is there latent demand for gas (either from industrial consumers or from other IPPs) which would mean that the gas import facilities would serve more than one buyer? If gas demand is very high, an FSRU solution may be too small-scale to meet that demand. The more diverse the sources of demand for gas, the less the project is dependent on a single source of demand (and, therefore, ultimately revenue).
- Regulatory requirements: do local regulatory requirements permit the same person to own and operate the relevant gas and power generation facilities, or indeed the cross- collateralisation of the facilities’ assets, in support of the other? Are there “open access” requirements to relevant utility infrastructure?
- Tax considerations: are there any tax considerations that would shape the structure? For example, are there withholding taxes that would prevent the on-lending of funds from one project to another?
- Timing: how urgent is the demand for incremental energy supply? The FSRU solution typically offers the prospect of quicker project delivery than a land-based terminal.
Taking account of the above considerations, we set out below some of the possible structures.
Option 1: an integrated model with a single project vehicle
This is probably the simplest, although in a sense least flexible, model. Under this model, a single project vehicle develops and constructs both the gas and power infrastructure, and raises the funding to do so under a single financing.
Option 2: a regas tolling model with separate project vehicles
Under a regas tolling model, separate project vehicles develop, construct and own the gas and power infrastructure. The gas is purchased by the power company direct from LNG suppliers on the market and the LNG is then stored and regasified by the gas company under a tolling arrangement. Each project company may raise its own financing or the funds may be raised under a single financing (with both the power company and the gas company acting as borrowers).
Option 3: a gas sales model
Under the gas sales model, separate project vehicles develop, construct and own the gas and power infrastructure. However, the LNG is purchased by the gas company and on-sold as gas to the power company. The power company may be only one of a number of purchasers of gas. The financings will likely be separate and indeed there may be a degree of government ownership in the gas company if the gas is to be sold to multiple end-users.
As stated above, a number of factors will influence the eventual choice of contract model. Option 1 favours smaller and simpler projects, where there is no real likelihood of third party gas sales and no issue with common ownership and operation of gas and power facilities. Option 3 is a more flexible model which may see the greatest degree of independence between the gas and power facilities, with the IPP perhaps acting as an anchor purchaser of gas but with a firm intention of the gas company to sell gas to multiple parties.
Option 2 represents perhaps the middle ground, recognising the reality that the gas facilities are possibly being developed primarily to serve the power project, but that there are good reasons to preserve some overall flexibility within the structure. It is option 2 that we analyse in more detail below.
Regas tolling structure
We set out in figure 2 a possible structure chart for an LNG to power project developed on the basis of a regas tolling structure (excluding finance documents):
Figure 2: Possible structure chart for an LNG to power project developed on the basis of a regas tolling structure (excluding finance documents)
As the chart shows, the project will be undertaken by two project vehicles: one developing, owning and operating the gas infrastructure (GasCo) and the other developing owning and operating the power plant (IPPCo).
As mentioned above, the gas infrastructure could be developed on a number of different bases: essentially, all the LNG import, storage and regasification infrastructure could be land-based. The alternative approach would see a minimum of land-based infrastructure, with the storage infrastructure located on a vessel and the regasification equipment located either on a vessel or a jetty.
IPPCo would undertake the project very much on classical IPP lines, underpinned by a long-term availability-based PPA with a creditworthy off-taker. However, instead of purchasing its gas under a long-term gas supply agreement or structuring its PPA as a tolling agreement with the off- taker supplying the gas and purchasing the resulting electricity, IPPCo in fact enters into one or more LNG sale and purchase agreements. IPPCo then tolls the LNG through the gas facilities under the terminal usage agreement (TUA). Whereas the PPA acts as the primary source of revenue for IPPCo, it is the TUA that underpins the economics of the gas facilities, and the charging structure under that agreement may well be on-regulated to address any "open access" requirements in the jurisdiction.
In relation to a classic gas-fired IPP with a single off-take, the project vehicle is remunerated via an availability-based capacity charge that will cover its fixed costs of constructing and operating the IPP and the shareholder's equity return. The variable costs of generation – principally, fuel and variable operation and maintenance costs – are paid for through a separate energy charge as and when the plant is dispatched. If the IPP is developed on a tolling model, the issue of matching the gas supply to power off-take does not arise. If there is a separate gas supply agreement, then either the contract must have no minimum annual contract quantity (ACQ) or, if there is an ACQ, then the costs of IPPCo failing to purchase the ACQ must either be mitigated in the gas supply agreement (via make-up provisions or something similar) or passed through to the off-taker under the PPA.
Clearly, in relation to an LNG to power project, the economic structuring is more complex and the TUA and the LNG SPA both need to be taken into consideration. Like the PPA, the TUA will contain some form of fixed payment, perhaps a capacity fee payable by IPPCo that will reserve an element of the terminal's capacity for the IPP and cover the fixed costs of the procurement/construction of the gas infrastructure, with a variable charge covering the variable costs of tolling any LNG through the facility. Equally, the LNG SPA with the LNG supplier will be structured on a "take-or-pay" basis with a minimum ACQ requirement (although there may be some ability to mitigate the LNG payment obligations: see below). Both the fixed costs of the TUA and take-or-pay nature of the LNG SPA will need to be taken into account in the PPA tariff. Clearly, the IPP needs to be predicated on the basis of reasonably high levels of plant dispatch to make economic sense.
As mentioned earlier, there is also the question of "project-on-project" risk. A delay in the construction of the gas facilities will leave the IPP unable to generate, and similarly a delay in the construction of the IPP will mean the gas facility will stand idle. In both cases, IPPCo will be unable to take delivery of LNG under the LNG SPA, and will potentially be incurring take-or-pay liabilities under the LNG SPA (unless it has, for example, been able to delay the commercial start date). On a typical IPP, construction delay risk would reside wholly with the EPC contractor, whose delay liquidated damages would be set at a level to keep debt and equity whole (subject to appropriate caps).
However, in relation to an LNG to power project with a minimum of two EPC contracts (one for the gas facilities and the other for the power plant) plus possibly a charter arrangement for the FSRU, it is not typically going to be commercially feasible for a delayed contractor under one contract to assume liability for loss of revenue across the entire project. Project sponsors may have to consider other mitigants. These may include:
- careful inter-project scheduling to allow a suitable buffer between scheduled commissioning of the upstream (and possibly cheaper) gas terminal and the scheduled commissioning of the downstream (and possibly more expensive) power plant;
- seeking as much flexibility for IPPCo as LNG buyer as possible around the timing of the commercial start date (including some controls for IPPCo over the "windowing" mechanism to determine the commercial start date) and the initial contractual volumes. Whether the LNG supplier will cede this degree of control will
- depend partly on its core business model. If the LNG supplier is a greenfield liquefaction project going through its own commissioning phase at the same time, it will likely want to retain control of the commercial start date under the LNG SPA; whereas if it is a portfolio supplier, it may be able to take a more flexible approach. As regards volume, IPPCo (as LNG buyer) may be able to negotiate reduced ACQ during an initial ramp-up period and downward quantity flexibility in case delay reduces its demand for LNG;
- in the case of a FSRU solution, maximising the contractual flexibility to defer the FSRU's delivery date and to deploying the FSRU as an LNG carrier to generate alternative revenue pending completion of the power plant;
- ensuring the power plant has a suitable back-up fuel (diesel) storage and unloading capability (although this will inevitably be limited); and/or
- sizing any contingent equity (and debt?) to provide a risk buffer for these events.
Similar project-on-project issues may arise during operations. For example, a forced outage affecting one piece of infrastructure may affect the performance of other infrastructure within the integrated project.
With two separate project vehicles owning the gas and power infrastructure, and with separate construction and operation arrangements, there are a variety of options that could be considered for the financing:
- Separate financings
Each project could raise finance separately and enter into its own facility agreement with different lender groups.
- HoldCo financing
There could be a single "HoldCo" financing with debt pushed down into each of the subsidiary project vehicles.
- On-lending structure
A single financing, where perhaps the more capex-intensive IPP company could borrow from lenders and on-lend a portion to the gas company.
- Single financing; dual borrowers
A single financing but with each of the gas company and the IPP company acting as borrowers and jointly and severally liable.
There is no "one-size-fits-all" financing solution for LNG to power projects, and the structure of the financing will need to address the requirements of the specific project, but the following points should be borne in mind:
- While two financings may permit a degree of flexibility (for example, with regard to subsequent separate refinancings and/or being able to allow each lender to opt to fund the asset they are most interested in), there will inevitably be intercreditor arrangements and potentially cross- collateralisation that will tie the two bank groups closely together. Also, there is the general execution risk of needing to close two financings at the same time. The effect on pricing of raising two smaller facilities rather than a single larger facility will need to be considered.
- In relation to the HoldCo and IPP on-lending structures, consideration will need to be given to the tax consequences of the on-lending structure, as well as the potential complexities of up-streaming debt service to lenders and available cash to shareholders.
- The joint and several structure will make the smaller GasCo jointly and severally liable for the debts of the larger IPPCo, and may introduce complexity if there are different shareholder groups for each asset and one of them wishes to sell down. There is also a level of complexity in relation to a dual borrower structure over and above a single HoldCo financing.
There is significant interest in LNG to power projects as a means of meeting demand for gas-fired power generation in many markets. The industry has delivered a number of successful LNG to power projects, and many others are at different stages of development.
An LNG to power project can be structured on several different models, depending on the specific characteristics of the project in question. Careful consideration of the project documentation will be necessary to ensure that risks are appropriately allocated, given the large number of project participants likely to be involved. Likewise, there are a number of different financing structures which may be available, depending principally on the composition of the sponsor group(s) and the appetite of the relevant bank market.