“On top of old smokey, all covered in snow”
American folk song
Snowy Hydro commissions Marsden Jacob to justify Snowy 2.0
Snowy Hydro recently released a report they commissioned from Marsden Jacob Associates “NEM outlook and Snowy 2.o – Marsden Jacob Final Report.
Somehow I find the word “final” amusing especially when applied to views about electricity.
The report seems to be essentially a marketing report that explains to politicians what electricity is, how the Australian system works and why investing in Snowy 2.0 is a good idea. All in 133 pages plus appendices.
As we have said many times when looking at such reports you can normally predict the conclusion by looking at who paid for the report. So it was no surprise to find this report says that doing Snowy 2.0 will add $4-7 billion of value to someone, not necessarily of course Snowy shareholders.
Whatever, this seems to be more of an economics report than a traditional modelling report. However, it’s hard to tell because of the vast amount of ground it tries to cover.
The report wants to cover the history and future development of the NEM, presents a brief didactic section on power economics (probably written more for the authors’ own benefit than that of the hopefully informed reader).
It also attempts to forecast what’s going to happen to dispatchable electricity and demand in Australia under no less than two scenarios and calculate the benefits from developing Snowy 2.0 as they seem to be in Marsden Jacob’s view of the world.
This is certainly ambitious.
On page 22 we learn that Snowy 2 is to be developed because:
“It is intended to address many of the challenges currently facing the NEM, which include firming intermittent generation, supporting the economics of existing coal-fired generation, and enhancing competition in the electricity value chain.”
It’s an interesting question as to whether we do want to support the economics of existing coal generation. I am fairly sure many readers think that we need to close those power stations as fast as can possibly be managed subject to keeping a stable power system in action.
What we can say for sure is it’s not Snowy’s business to be worrying about coal genenerators. It doesn’t own any such generation. Its job is to earn a return on capital subject to its social constraints.
On the positive side
We expect that discussion about how much and what form dispatchable generation is required is going to occupy the electricity sector stakeholders more and more.
This report is one of many that note the coal fleet is going to have to be replaced and that heavily based VRE (variable renewable energy) grids will need some dispatchable energy.
The report also correctly notes, not that it is in the slightest bit new information, that ramping up and down has a negative impact on the economics of coal fired generation, and for that matter combined cycle gas.
Still, the idea that only Snowy 2.0 can solve the problem seems, to put it mildly, self serving.
Modelling discussion takes 3 paragraphs
The entire discussion of the modelling methodology and results takes 3 paragraphs (in a 150-page report).
Snowy 2.0 would provide market benefits that reflect a reduction in capital and operating costs (mainly fuel costs) that would otherwise be needed for the production of wholesale electricity and maintaining supply reliability in the NEM.
From the two scenarios modelled, the table shows the present value of the market benefit from Snowy 2.0 operation excluding the market benefit of the option, the market benefit of the option provided by Snowy 2.0, and the total. The present values were determined over the period from 1 July 2018 to 1 July 2074 at a real discount rate of 4.55 per cent.
The market benefits, excluding optionality, is composed of avoided capital cost, energy arbitrage (including drought conditions) and FCAS.
The report states that the “benefits” of developing Snowy 2.0 in a Net present value sense range between $4.4 bn to $6.8 bn. This is a wide range but the report’s not-too-implicit message is mainly: it’s a big number and therefore the reader can safely conclude Snowy 2.0 should be developed pronto.
We would like to see more discussion of the modelling methodology beyond the fact that two scenarios were developed each with and without Snowy 2.0, we pressed the button and here are the results. Thanks for coming. On to our next gig.
Nothing but Snowy 2.0 will do
One of our main immediate criticisms of the paper is its assumption that Snowy 2.0 is the only pumped hydro project capable of being developed in Australia.
For instance as early as page 5 we find the following passage:
“Market development without Snowy 2.0
For the two renewable generation development scenarios, we modelled the NEM on the assumption Snowy 2.0 is not developed. We found as follows:
intermittency issues will need to be addressed using batteries, peaking generators and coal-fired generators. Batteries will be developed to support the fast response required under 5-minute pricing and to contribute to firming intermittent generation in retailers’ portfolios. However open-cycle gas turbine (OCGT) plant is mostly used in this role
without large-scale storage, no additional renewable generation above the minimum targets will be economic. In the current policy-only scenario (that is, the LRET+VRET scenario), the LRET and VRET are satisfied and no more.”
Your author believes that there are many other pumped hydro possibilities in Australia besides Snowy 2 and some of them may well have better economics.
At least three sites are under active development right now. Not only that, there are a variety of other technologies, admittedly less technically or economically proved, than batteries or pumped hydro that might be capable of doing the job.
The statement that no additional renewable generation will be “economic” is bold.
Of interest in the study is that external benefits are noted. These are benefits such as reduced fuel costs of say gas generators that are avoided by Snowy development.
Of course Snowy can’t capture those benefits. But the report treats them as “market benefits” using the RIT-T investment test for transmission.
We have seen sporadic attempts to measure externalities before. For instance the ground breaking Massachusetts “State of Charge” report did exactly that when looking at the economic case for batteries. We do think such externalities are worth considering, but it’s an areas to consider closely.
For instance, if you do it for pumped hydro you also need to do it for, say, behind-the-meter batteries and look at all the transmission and distribution economics.
In any event traditionally we don’t in a private enterprise system look at externalities, either benefits or costs.
The project developer has to look at revenues and costs they face. It’s up to the system provider to provide incentives eg REC prices or costs eg a carbon price for perceived externalities and make them relevant to the developer.
Startling conclusions about impact of 5 minute settlement
Readers have to skim another 50 pages of unnecessary background to get to the next “insight”. At an investment bank you’d be in trouble for making time poor clients read all the “fluff” in this report.
At page 75 we get to a section which no doubt John Piecre (AEMC chairman) will be emailing to staff, headed “Future of NEM Reforms – 5 minute pricing”
The section states:
“Given the inability of existing peaking plant to respond to 5-minute prices from a cold start, there would be a vast reduction in the level of cap contracts that can be supported by currently registered generation. In fact, existing plant might not be able to provide any cap contracts …..
It is estimated that, due to the technical capability of peaking plant, the amount of caps available from existing providers under a 5-minute settlement market would decline from the present level by at least 4,000 MW, and most probably substantially more……
This would mean that, without alternative means of covering cap contracts, the suppliers of cap contracts would need to keep plant operating, including peaking plant, more than would have otherwise been the case. With more high-cost gas-fired plant operating under 5-minute dispatch, overall generation costs would be likely to rise, and increased gas use would further limit the availability of gas for other users (such as domestic and industrial users).”
The report goes on to state that OCGT coupled with batteries or gas fuelled reciprocating engines can provide cap contracts.
The report states that AGL’s 210MW, $295 million Barker Inlet station in south Australia has been delayed, however we note that an EPC contract has been announced AGL Barker Inlet plant EPC contract and engines are to be delivered by the end of this year.
It’s not clear to us that electricity consumers will be worse off, or that costs will rise as a result of 5 minute settlement. We will leave it that for now.
As for FCAS the report claims that 5 minute settlement will adversely impact FCAS prices. We think most observers think that FCAS prices are more likely to fall than rise as batteries take over that role and that there is already early data to that effect in South Australia.
David Leitch is principal of ITK. He was formerly a Utility Analyst for leading investment banks over the past 30 years. The views expressed are his own. Please note our new section, Energy Markets, which will include analysis from Leitch on the energy markets and broader energy issues. And also note our live generation widget, and the APVI solar contribution.