Solar Flagship Program Submission to the Boston Consulting Group

Solar Flagship Program Business Case: Victorian Government Submission Solar Flagship Program: Submission to the Boston Consulting Group Victorian Department of Primary Industries 6 August 2009 1 Solar Flagship Program Business Case: Victorian Government Submission Executive Summary The Solar Flagship Program (SFP) should be used to drive utilisation of Australia’s abundant solar resource to provide affordable, secure, reliable and sustainable electricity in a carbon-constrained world. The Victorian Department of Primary Industries (DPI) believes this is most likely to eventuate based on: • a clear statement of program objectives; • a clear distinction between program-wide objectives and targets, as opposed to legally binding deliverables for individual projects; • the use of output—based deliverables for funding purposes (gigawatt hours delivered); • the use of long-term cost-effectiveness as the predominant selection criterion; • adopting a flexible approach that does not overly prescribe the solution – particularly with respect to technology type, location, number of projects or number of operators; • support for a pipeline of projects; • avoiding the creation of perverse incentives; • application of the principle of funding source neutrality; • transparent and comprehensive costing; • maximum possible consistency with state/territory initiatives (where relevant); and • employment of selection criteria to assess the comparative benefits of other project attributes (including research; employment of storage and hybrid technologies; and co-location/solar parks). Introduction The Victorian Department of Primary Industries (DPI) welcomes the opportunity to comment on the potential design of the Solar Flagship Program (SFP). The SFP provides Victoria with an opportunity to secure a funding partner for the Victorian Large Scale Solar Project (VLSSP). For SFP and VLSSP to complement each other, however, it is essential to develop comparable program design and timetables. This submission provides a background to the VLSSP. It then addresses the specific features of the SFP that will have an impact on the compatibility of the two programs. It then makes recommendations on DPI’s view as to the optimal design of these features. Background In June 2008 the Premiers of Victoria and Queensland visited solar electricity generation projects in the United States. The visit made clear to both Premiers the potential to better utilise Australia’s solar resource. Following this visit, the Premier of Victoria announced, on 10 March 2009, that Victoria would provide up to $100 million, net present value, for the development of a Victorian-based facility capable of producing 330 gigawatt hours (GWh) per annum of electricity from solar energy. A Request for Proposals (RFP) document was 2 Solar Flagship Program Business Case: Victorian Government Submission released on 9 April 2009, followed by a stakeholder forum on 15 May 2009. Details of both the RFP and stakeholder forum are available at www.dpi.vic.gov.au/largescalesolar. Applications for the Victorian RFP close on 25 September, with assessment to take place by the end of 2009. It is anticipated that an announcement will be made as to a successful project at that time. The Victorian RFP makes clear that proponents are expected to seek Commonwealth Government co-funding. As this is likely to come from the SFP, it is in the interests of both programs that key design details align as best as possible. Program Objectives and Principles The final business plan should be prefaced with a clear statement of the objectives sought by the SFP. This would serve to guide a series of subsequent decisions – including program targets, selection criteria, timing and governance. The primary measure of success of any expenditure of public money should be efficiency – that is, obtaining the maximum desired outputs per dollar spent. Consequently, once the decision has been made to provide public money to encourage electricity generation from solar energy, the logical conclusion is that the program objective should be to achieve the maximum quantum of electricity generated for each dollar of public money spent. Where a program is designed to commercialise new technologies, this assessment needs to be made against the potential of supported technologies to generate electricity at a given cost over the long term – a point discussed further below. DPI believes that a similar objective should be the basis for the SFP. On this basis, the following comments are made. Recommendation: That the objective for the SFP should be clearly articulated and long-term cost-effectiveness (levelised cost of electricity produced by a technology, over the long term) should be the predominant objective of the SFP. Program Targets The 2009/10 Commonwealth Budget and subsequent documents indicate that the SFP will comprise the following: • 1,000MW of built solar capacity; • Mixture of photovoltaic and concentrated solar thermal power technologies; • Inclusion of storage; and • Inclusion of an R&D component. The final business plan for SFP should make a clear distinction between: • program targets; • assessment criteria; and • the basis for payment milestones. 3 Solar Flagship Program Business Case: Victorian Government Submission Program targets are to some extent aspirational, and may not be legally binding on an individual party. DPI does not have a view as to how the Australian Government wishes to articulate such targets, other than to state that they should be consistent with the overarching principle suggested above – as well as existing policy measures at all levels of Government. Of more interest to DPI are payment milestones, as these will constitute the deliverables which are legally binding on individual project proponents. In DPI’s view SFP should structure these according to delivery of units of electricity produced from solar energy (GWh/a). This would ensure that public money actually achieves the objectives of the program by linking it directly to useful outputs. It would also serve to protect public money from risks associated with non-delivery by a project proponent, as payments could be provided at milestones which closely mirror actual electricity production. While it could be argued that a capacity-based funding milestone(s) would be easier to administer (a single disbursement could be provided at time of construction), this could potentially lead to the construction, at considerable public expense, of a facility which ultimately produces less power than expected – with less public benefit. In brief, capacity-based funding does not correctly incentivise parties to behave in accordance with the suggested objective above, and produce as much power from solar energy as possible at least cost. Other desired program outcomes, such as technology mix, inclusion of storage and research and development, should be addressed as qualitative selection criteria. This allows for a considered decision as to what trade-offs will be required in awarding funding to a given party. This is discussed further below. Recommendation: The SFP should distinguish between program-wide targets and project deliverables which will be legally binding on individual parties. These later deliverables should be based on actual delivered electrical output (GWh/a). Technology neutrality In DPI’s view the SFP should be technology neutral, within the overall confines of seeking a given quantum of electricity generation from solar energy. The current SFP indicates that there will be a 50/50 split in awarding projects between concentrated solar thermal and solar photovoltaic technologies. While this may ultimately be the outcome of a competitive, technology-neutral process, it should neither be anticipated, nor required, that this precise split will eventuate. Rather, a strict adherence to technology neutrality should prevail. This approach is most likely to satisfy the objective of cost-effectiveness. On the assumption that the principle of technology neutrality is adopted, it should be noted that other thresholds (such as size, location etc) may discriminate in favour of one technology versus another. For example, solar thermal technologies rely on the use of boilers which are only scaleable in a “coarse” or “granular” sense. Photovoltaic arrays, however, are highly scaleable technologies, which are also wellsuited to distributed applications. 4 Solar Flagship Program Business Case: Victorian Government Submission Recommendation: The SFP should be strictly technology neutral. Inclusion of program criteria should be avoided unless they can clearly demonstrate that they: • Are necessary to deliver additional benefits/avoid disbenefits; and • Will not corrupt the principle of technology neutrality. Number of power stations Currently SFP documents indicate that four (4) electricity generators will be built, totalling 1,000MW. This implies four generators of approximately 250MW each. Consultation to date between DPI officers and lenders (domestic and overseas merchant banks and venture capitalists) indicates that there is significant lending risk attached to the construction of a solar generator of this size. In the last decade the largest solar electricity generator constructed has been Nevada Solar One at 64MW (parabolic trough). Other technologies (heliostat field CSP/PV; parabolic dish PV/Sterling Engine; flat plate photovoltaic etc.) have not been built to a scale this large. Consequently for most lenders the scale-up to 250MW is challenging, with many indicating that a smaller figure would be more commercially attractive. If the four generator target was relaxed as a program criterion, it would allow for the receipt of a larger number of proposals of a smaller size. Investors have indicated to DPI that 10 projects of 100MW size would be more manageable. This would yield multiple benefits, including: • Lower financial risk (leading to an increased likelihood of funding, lower capital cost and consequently lower electricity costs) • Better capacity to distribute supply augmentation across the grid • Better capacity of market to absorb new supply increments • Better capacity to assist a broader range of technologies to market • Better capacity to leverage state/territory support (financial and non-financial) A larger number of smaller generators may also yield benefits in terms of construction timelines – an issue discussed further below. To the extent that larger sized generators can benefit from economies of scale, allowance should be made for projects to be staged over a number of years. The concept of solar parks (discussed below) may be relevant to this point. Recommendation: The SFP should be flexible in its approach, rather than prescriptive, with regard to the number of projects funded, and allow for a larger number of smaller projects, either standalone or as stages in a larger project, as determined by individual investor proposals. Location (Solar parks) In general, there are two types of costs associated with a new solar electricity generator: 5 Solar Flagship Program Business Case: Victorian Government Submission • • Those which are largely scalable, based on the size of the plant (MW; GWh/a); and Those which are less scalable. Some examples of relatively non-scalable costs are expected to include roading, network augmentations, as well as environmental and land use approvals. Other costs (land acquisition, balance-of-plant infrastructure) may also become proportionately less where larger volumes of electricity are generated from a single location. Where costs are relatively non-scalable, they suggest efficiencies to be realised from either: • Construction of single, larger plant; or • The sharing of services by a multitude of physically co-located smaller plants. Ultimately location choice will be driven by the need to optimise for a range of costdrivers, including solar resource, land costs, network connection costs, network loss factors, logistics costs, water availability (including quality and cost), availability of a local labour force and, where relevant, availability of fuel (for example, gas co-firing). These are expected to be determined by the proponent’s technology choice. Consequently, if the SFP is to be technology neutral (as is argued above), individual project proponents should be free to determine which arrangement best suits their business case. Where a shared service/solar parks arrangement would offer a particular project some benefit, the relevant project proponent should, in their proposal: • quantify the extent of these benefits; • quantify the extent of any risks associated with the provision of such shared services by all affected parties (i.e. in the event that they are not fully subscribed by subsequent market entrants); and • explain how these translate into a lower levelised cost of electricity for consumers1. It should also be noted that it may ultimately prove more economically efficient to locate a larger number of smaller plants across different areas of the electricity network. Recommendation: • SFP should be flexible with respect to location. Projects should not be required to site themselves in a particular area in order to be eligible for consideration under the SFP; and • Where a project proponent proposes co-location with other generators in an area offering shared services, all costs associated with the provision of those 1 The Australian Energy Market Commission (AEMC), in its Second Interim Report on Energy Market Frameworks in Light of Climate Change (June 2009), recommends the establishment of Network Extensions for Remote Generators (NERGs). NERGs would allow for pioneer network infrastructure to be built to areas identified as having a viable renewable resource; the costs of this infrastructure would, in the first instance, be borne by electricity consumers, until such point as the new network capacity is subscribed by generators. 6 Solar Flagship Program Business Case: Victorian Government Submission services should be apportioned to that project for purposes of assessing its cost-effectiveness ($/MWh). Number of system operators The current SFP documentation indicates that a single operator will be chosen for all solar electricity generators constructed. Victoria does not support this approach. It is not clear what benefit this approach brings. On the contrary, this approach is more likely to bring a disbenefit in the form of reduced competition, and is unlikely to be in the interests of electricity market participants. In the current electricity market, a number of retailers are currently acquiring supply assets – including renewable generation assets - as a natural hedge against market uncertainty (both REC and “black” electricity). Such retailers would be expected to have an interest in acquiring solar assets for the same reason. Consultation between Victoria and the retail sector confirms this is in fact the case. It should also be noted that it may be easier for multiple proponents to secure funding based on multiple balance sheets rather than one. Recommendation: SFP should be flexible in its approach to asset ownership. No artificial impediment should be constructed which limits ownership diversity in the emerging solar portfolio. Avoid perverse incentives The structure of SFP funding criteria is essential to avoid perverse incentives. An example is furnished by Spanish feed-in-tariffs. These provide a generous revenue stream (equivalent to approximately AUS $500/MWh) for output from a solar plant with a steam turbine rated at less than 50MW. Industry has consequently attempted to maximise revenue within these parameters. This has led to construction of multiple <50MW units, with a storage component, in order to maximise the output (MWhs) attributable to the plant. Conversely, there has been no incentive to reduce costs to customers. Systems of this type are inefficient for several reasons: • Smaller turbines are less thermally efficient than larger units would be; and • The quantum (and cost) of stored power is not justified on the basis of meeting system energy (demand). Recommendation: Any Australian program should only incentivise cost effectiveness, not specific size thresholds, configurations, locations or technology types. Research component Current documentation for SFP indicates that projects must incorporate a research component. This appears to be poorly justified, as it requires a project proponent to 7 Solar Flagship Program Business Case: Victorian Government Submission simultaneously satisfy criteria for both commercial viability, as well as innovation. This is unlikely to be practicable, and is likely to lead to the artificial – or perhaps tokenistic – incorporation of an unrelated research task to what is primarily a commercial enterprise. To the extent that Government wishes to foster innovation at the early stages of the research-development-demonstration chain, it should focus programs on those specific stages. This would allow programs to be crafted which adequately match the different technical and commercial risk profiles associated with these distinct stages (e.g. high technical risk, low commercial risk at the research stage; lower technical risk, higher commercial risk at the demonstration stage; negligible technical risk at the commercialisation stage). This approach has characterised Government programs for renewable energy in the past. For example, the Victorian Energy Technology Innovation Strategy (ETIS) provides grants for research and development, as well as large-scale demonstrations of pre-commercial technologies. It is not an expectation of this initiative that any constructed plant will be a viable commercial entity on a long-term basis; rather the chief criterion for success under these programs is whether a concept is proven. Industry has also indicated to the Victorian Government its support for a clear, simple goal for the proposals. Any complexity (such as the need for a significant R&D component) will put the SFP at a disadvantage compared to other incentive programs overseas. Recommendation: Incorporation of a research component should not be a requirement for projects under the SFP. Neutrality applied to funding partners SFP projects are expected to draw co-funding from both the private sector as well as regional (state, territory or local) governments. Victoria has declared its intention to partner with the Commonwealth to the sum of $100 million through the Victorian Large Scale Solar Project. Other jurisdictions may adopt a similar approach. Prior Australian Government renewable energy programs have differentiated between state/territory and private sector funding for purposes of fund-matching/leverage criteria. For example, the Renewable Energy Demonstration Program guidelines stated: “Applicants cannot use funds obtained from State, Territory, local government or other Australian Government grants or financial contributions as matching funds. Contributions from state or local government grants are welcome but would be considered additional to the two for one ratio of applicant to Australian Government funding.2” Renewable Energy Demonstration Program: Information Guide. Department of Resources, Energy and Tourism, 2009, p. 7 2 8 Solar Flagship Program Business Case: Victorian Government Submission The SFP should not discriminate between funding sources on the basis of them being private sector or government. This applies regardless of the form of the financial contribution (grant, equity ownership, loan, offtake agreement etc). Such discrimination acts as a disincentive for State and Territory Governments to share in the risks of such ventures, and acts as a disincentive for private investors to seek such funding. It may also be difficult to distinguish in practice, as many “private” entities investing in such a project may be wholly or partially owned by Australian or overseas governments, or only capable of providing particular terms of finance due to arrangements with a government (loan guarantees etc). To the extent that a proposal is contingent upon receipt of government funding through another competitive process (such as Victoria’s), this contingency should be viewed in the same light as a commercial contingency (i.e. private finance is also expected to be subject to conditions precedent). Recommendation: State or other government contributions to a project should be treated the same as private sector contributions. Transparent and comprehensive costing If – as recommended above - the main selection criteria for SFP projects is costeffectiveness, it will be essential that levelised costs of electricity generation are based on full and transparent costings. This should include costing of risks to third parties (such as non-subscription of shared services as described above) as well as other forms of government assistance (provision of free land; tax concessions; loan guarantees etc). Recommendation: Levelised costs of electricity generation for a project should be based on transparent and comprehensive costing. Prior assessment Wherever there is a case of multi-jurisdictional funding, alignment of assessment processes (state/territory and Commonwealth) would be optimal. This would ensure efficient use of time and money, as well as consistency. This may not always be possible. To the extent that some independent assessment of a project has already occurred (i.e. the proposed Victorian process) the information obtained from this assessment should be shared with the Commonwealth, and should be taken into consideration by the Commonwealth in making its own determinations under the SFP. Similarly, there should be sufficient flexibility in the SFP to accommodate regional differences and requirements. Victoria has established a process which has been accepted by stakeholders; other jurisdictions may also elect to pursue incentive schemes for solar energy. To the greatest extent possible, SFP should try to work in collaboration with such jurisdictional approaches, where they exist. 9 Solar Flagship Program Business Case: Victorian Government Submission Recommendation: That the SFP be designed to work collaboratively with any jurisdictional incentive schemes, to the greatest extent possible. Storage The SFP Business Case Discussion Paper indicates that “the Government proposes that storage be a part of all Flagship projects”. As with the other issues, this should not be a threshold criteria (cause for prima facie exclusion of a proposal). Rather, capacity of a plant to contribute to system stability and supply security (through such measures as storage) should be included as one of several assessment criteria. Storage is an expensive addition to a plant. Its inclusion must be justified based on some market/consumer benefit. One example would be where the stored energy will command sufficient revenue from the wholesale electricity market to justify the increased upfront capital expenditure, increased land area required and the increased technology risk. Another may exist where the location of a large scale solar power generator would otherwise put regional grids at risk of instability due to intermittency of generation. In addition to storage technologies, the incorporation of other non-solar elements should be assessed on their intrinsic merits. For example, gaining the highest possible capacity factors from a generation asset makes economic sense and should be encouraged. Consequently, SFP should allow proponents of solar thermal plant to employ hybrid systems (e.g. gas co-firing, geothermal co-heating) in order to fully utilise turbines and balance-of-plant when the solar resource is insufficient. Recommendation: Incorporation of storage should not be a mandatory condition of project funding under the SFP. Rather, the benefits of incorporated non-solar elements – including storage and hybrid systems - should be assessed on a case by case basis through the SFP assessment criteria. Least cost generation – short term versus long term The SFP Business Model Design – Request for Submissions (July 2009) posed the question as to whether weighting should be given to a technology which, while not the least-cost solar technology at the moment, held potential for strong future cost reductions. The basis of providing targeted assistance to solar technologies is the belief that this suite of technologies has significant potential to provide secure, affordable, reliable and sustainable electricity for Australia – and the World – over the long term. Consequently, as discussed at the beginning of this submission, DPI believes the objective of the SFP should be stated as the delivery of lowest levelised cost of electricity from solar over the long term. The challenge for the Australian Government – and its project partners – will be apportioning this broader program objective to individual projects in a legally binding manner. This may, in a given round, necessitate funding a project which does not have the lowest levelised cost of electricity generation in the short term. Where this is the 10 Solar Flagship Program Business Case: Victorian Government Submission case, Government should justify this decision by reference to detailed estimates of the cost components of the specific technology in question. This highlights the need to employ highly competent commercial and technical assessors. These parties will need to apply uniform and transparent assumptions as to project cost components and escalation factors. Even in the face of informed estimates of future costs, there is a risk that a technology investment made today will be eclipsed by a more cost-effective future technology. In the energy sector the issue is compounded by the (generally) large size of the assets. Consequently this risk could be mitigated by favouring the addition of smaller, rather than larger, generation assets. Such smaller assets commit a smaller amount of capital to a given technology. This further supports the merits of allowing SFP to fund a larger number of smaller projects, as discussed earlier in this submission. Recommendation: The SFP should seek to achieve lowest generation costs over the long term. Where this necessitates awarding funding to a project which is not least cost, in the short term, this decision must be robustly and transparently justified. Timelines for delivery Current SFP documentation indicates that the construction of all Solar Flagship generators is expected to commence in 2012, with commissioning from 2015. This construction schedule would essentially lead to an initial “spike” in activity, followed – in all likelihood – by a sharp contraction in activity. This would be expected to force different proposals to compete for the same limited pool of: • investment capital; • skilled staff; • components; and • offtake agreements with wholesale purchasers of electricity. Such a “boom-bust” solar industry would be expected to experience higher costs due to upwards price pressure on wages, materials and capital, while simultaneously depressing wholesale electricity prices. A boom-bust industry may also, in practice, not be attractive to young market entrants, who will be required to choose between a career in different sectors – some of which may offer greater prospects for job security. In addition, it is questionable whether overseas financiers, IP owners or plant operators will have an interest in developing an Australian presence when there appears little likelihood for long-term future growth. Consequently there may be merit in staging the SFP so that few proposals are commencing construction at a given time. This would allow for a “pipeline” of development over a longer period of time (i.e. 10 years or more) which is more likely to suit the finance community, the electricity market, and the skilled professionals who will be required to build/operate these plants. Prior experience in other industry sectors indicates that manufacturing capacity is unlikely to eventuate unless there is at least 10 years of activity anticipated. 11 Solar Flagship Program Business Case: Victorian Government Submission This may necessitate holding several SFP application rounds, with only one or two “winners” in a given round. Such a staged approach may, however, yield the additional benefit of allowing for “learning by doing” with respect to selection criteria, contract settlement, governance and other program details. Recommendation: Employ SFP funding to provide a “pipeline” of solar electricity developments in Australia over a longer period of time. Allow plant operators to retain/sell RECs In the absence of funding criteria to the contrary, a solar electricity generator would be expected to earn revenue from both electricity sales, and sales of renewable energy certificates (RECs) under the expanded Renewable Energy Target (RET). This should be the case. If funding conditions under the SFP were to stipulate otherwise, (i.e. the surrender of RECs to a Commonwealth body; ineligibility under the RET) it would simply be shifting a greater portion of project finances onto either the SFP, private lenders or state/territory government funding sources. Recommendation: Allow plant operators to retain and sell RECs. Selection criteria Selection criteria for the SFP should be clearly communicated to program participants at the commencement of the tender/application process. There are a range of issues which Government will need to consider in making an assessment of worthy projects. Consequently it is advisable to grant the assessment panel(s) the latitude to make a considered judgement across all relevant issues. Conversely, it would be ill-advised to establish too many “threshold” criteria (cause for categorical rejection of an application). This will ensure that otherwise sound proposals need not be rejected outright due to the fact they fall outside an arbitrary category (location, size, technology type etc). By making such criteria part of a weighted list, the assessors will have maximum flexibility to consider applications on their merit across the full range of relevant issues. Recommendation: Employ as few threshold criteria as possible, relying instead on multi-criteria analysis to provide assessors with the capacity to make a considered decision across a range of criteria. 12