Unreliables are more about flopping than firming

Unreliables are more about flopping than firming.

WTF?

The author states:" It's fair to say that GenCost is not easily digested."

Yet on the first page of the CSIRO website we get:
"Key points

The report found renewables remain the lowest-cost new-build electricity generation technology, while nuclear small modular reactors (SMRs) are the most costly.
Electricity systems rely on a mix of technologies, because no single option can deliver all the capabilities required for a reliable, secure and flexible supply.
Rising construction costs in Australia and supply chain constraints for some technologies remain a challenge for reducing costs".

Probably hard to accept rather than hard to digest.

Later on I read: " According to ChatGPT....."

That won't go down well with the "do your own research " mob as they look to advice from crazies on youtube.

To WTF?
The 2025-2026 GenCost report by the CSIRO is deceptive as it deliberately avoids mentioning anything about the costs of renewables that are additional to the actual initial cost of building and operating wind and solar. But the 2024-2025 GenCost report does not try to hide these additional costs (although it doesn't give a $ figure), so look at figure 6-7 and read section 6.3 on pages 79 to 81 of the 2024-2025 report. These pages show that a renewables-based energy system must have back-up generating capacity at least equal to the total peak demand: in other words, spend tens or hundreds of billions on a renewable energy system capable of meeting peak energy demand when the sun is shining and the wind is blowing and then build a fossil fuel energy system of the same peak demand capacity to provide electricity for when the sun isn't shining and the wind isn't blowing. Here in Australia, we are still many billions of dollars away from completing out transitioning to a renewables-sourced electricity grid, so expect electricity prices to continuing to go up!

Bernie,

GenCost doesn't "hide" system costs, but its headline figures are often quoted as if generation cost equals system cost, which is not what the reports actually say. Reliability, backup, storage and integration are discussed in the body of the modelling, even if they're not collapsed into a single headline number.

The step from that observation to claiming that a renewables-based system requires building a full fossil system of equal peak capacity alongside it is not supported by the modelling. What's required is sufficient firm capacity, which can be delivered through a mix of storage, dispatchable plant, transmission, demand response and, during transition, gas. That mix is costly, but it is not the same thing as duplicating the existing fossil fleet.

Conflating firm capacity with a parallel fossil system overstates the case and muddies what should be a straightforward system-cost discussion.

To John Daysh
Can you explain why neither of the two most recent GenCost reports fail to give a dollar estimate of the cost of what you call firming capacity? The 2024-2025 report talks about the need for firming capacity and the 2025-2026 report makes no mention of it whatsoever. Why is the government hiding these costs from the public? Really, there's only one reason - because the costs are ginormous and would scare the living daylights out of taxpayers and users of electricity.

Indeed I can, Bernie.

GenCost doesn't put a single dollar figure on "firming" because there isn't a single thing called "firming" to price. It's not a standalone technology, it's a system property that emerges from different mixes of storage, dispatchable capacity, transmission, overbuild and demand response.

The cost depends entirely on the scenario being modelled.

That's why GenCost presents generation costs separately and then discusses reliability and system needs within scenario modelling rather than as a single headline number. A single "firming cost" would be meaningless without first specifying the mix, the penetration level, the reliability standard and the time horizon.

That's not the government "hiding" costs, it's the limits of aggregation. The costs are real and material, but they show up distributed across storage, networks and dispatchable capacity rather than as one scary line item. Collapsing them into a single dollar figure without those assumptions would be more misleading, not more transparent.

The legitimate critique isn't that these costs are secret, it's that public debate routinely ignores system costs altogether and fixates on generation-only headlines. But that's a communication failure, not evidence of a cover-up.

To John Daysh
If what you say is correct, then government is being dishonest or incompetent or both by not directing the CSIRO to come up with a detailed scenario of whatever energy production system they believe will be cheap, effective and reliable. By failing to give this direction to to CSIRO, they are trying to hide the total cost of Australia's future energy system in the hope that voters will remain ignorant and not chastise them at the ballot box when they (the voters) find out that billions of their taxpayer dollars could have been redirected into defense, housing, education, health or a multitude of other government services.

Your conclusion doesn't follow, Bernie.

GenCost isn't intended to be a single, definitive blueprint for "the" future energy system, and CSIRO isn't being directed to produce one because there is no single system to cost. Reliability, mix and total system cost are explored through multiple scenarios across AEMO's ISP and related modelling, not collapsed into a single politically convenient number.

What GenCost does is provide consistent, comparable input costs for the technologies. The amount of storage, backup, transmission or demand response you end up needing depends on how reliable you want the system to be and what risks you're willing to wear.

Those choices are political, not scientific, which is exactly why there isn't one tidy number to quote.

But that doesn't mean system costs are trivial or unimportant. It means they can't honestly be reduced to one "hidden" dollar figure without first specifying the system you want, the level of reliability you're prepared to pay for, and over what period of time. Pretending there's a single suppressed number conflates complexity with concealment.

So the problem isn't that voters are being kept ignorant of some secret total. It's that public debate routinely treats generation-only cost headlines as if they settle the question, when they plainly don't. Again, that's a failure of political communication (and media simplification), not evidence of a deliberate cover-up.

Bernie McMasters,
I have long suspected John Daysh of having invested in the failing present technology renewables industry & he is desperate to maintain the momentum of the Gravy Train !

Indyvidual
I agree. Daysh 'doth protest too much'. I am making a simple point: why does the government not direct the CSIRO or AEMO to devise the most likely/best scenario for Australia to reach Net Zero and then produce a costing for that scenario. Instead, Daysh seems to be denying the need for taxpayers to be told the ultimate final cost of Net Zero by saying it's political. Of course it's political and it's shameful in my view that the CSIRO and AEMO don't stand up to the political games being played on this issue and do what independent entities are supposed to do, namely, give fearless independent advice.

Bernie,

I haven't denied the need for system-level costing, and I haven't claimed it's illegitimate to ask what a Net Zero transition will cost. What I've said, twice now, is that there is no single "ultimate final cost" that can be produced independently of policy choices.

Asking CSIRO or AEMO to publish the cost of Net Zero presupposes agreement on reliability standards, risk tolerance, demand growth, technology mix, time horizon and financing assumptions. Those inputs are not scientific facts waiting to be discovered, they're choices that governments have to make.

Different choices produce different totals.

That's why these bodies model multiple scenarios rather than declare one "best" system and attach a single headline price tag to it. Presenting one number as the cost would be misleading unless all of those assumptions were fixed first, which is exactly why serious system modelling doesn't work that way.

None of that excuses oversimplified public messaging or generation-only headlines being treated as decisive. But complexity isn't concealment, and refusing to collapse it into a single political number isn't a failure of independence.

To John Daysh
and what I'm simply asking is that the government decide on what you call 'policy choices' so that a single ultimate final cost can be determined by the CSIRO and AEMO and then communicated to the people of Australia so that we know what it's going to cost us to achieve Net Zero.

and you seem to be defending the CSIRO and AEMO by saying they model multiple scenarios but they don't produce a costing for of the scenarios, so what are they hiding, I wonder? I suspect it's a cost of about one trillion dollars, with much of this amount to be paid by taxpayers, either through their purchase of domestic solar panels and batteries or through subsidies to industry or both.

Bernie,

It appears you've collapsed fully into conspiracies, motive-hunting, and citing invented numbers.

Governments do make policy choices, and those choices are then modelled in detail. That modelling doesn't happen in GenCost because GenCost isn't a system-design exercise, it's an input-cost study. Whole-of-system scenarios, including reliability standards and total system costs, are produced through AEMO's Integrated System Plan and related modelling, where assumptions are explicit and alternatives are compared side-by-side.

It's simply incorrect to say that scenarios aren't costed. They are. What doesn't exist is a single, immutable "ultimate final cost of Net Zero", because costs vary with demand growth, technology mix, reliability settings, fuel prices, financing assumptions and time horizon. Change those inputs and the total changes materially.

Again, presenting one number without locking all of that down would be misleading, not transparent.

Jumping from that complexity to claims that costs are being "hidden", or asserting a trillion-dollar figure without a model, assumptions or source, isn't analysis. It's speculation. If there's a specific scenario you think governments should commit to, the productive way forward is to specify its assumptions and cost that system, not to infer concealment from the absence of a single headline number.

To John Daysh
You wrote "GenCost isn't a system-design exercise, it's an input-cost study" yet Bowen happily quotes the GenCost reports, saying that renewables are the cheapest way to produce electricity. Why does the federal government not direct AEMO or whoever is responsible for energy-related studies to produce a design of the full system and then direct CSIRO to assess the cost of the complete package of input measures? Answer: because the cost will be a trillion dollars or more, as implied in the 2024-2025 GenCost report.
I'm not sure why you are trying to defend the indefensible.

Bernie,

You're still conflating three different things: political messaging, input-cost studies, and system-design modelling.

When Bowen quotes GenCost to say renewables are the cheapest generation technology, that's political messaging based on a generation-only metric. I've never defended that simplification, and I've explicitly said it's misleading when treated as a whole-system answer.

GenCost itself is not pretending to be a full system design, and it does not "imply" a trillion-dollar cost. It provides technology input costs. Full system design and costing is done through AEMO's Integrated System Plan and related scenario modelling, where assumptions are explicit and alternatives are compared.

That work already exists.

What you keep demanding is a single, authoritative "final cost of Net Zero" detached from assumptions about demand growth, reliability standards, technology mix, financing and time horizon. That number does not exist in any serious modelling framework, and inventing one while accusing others of concealment isn't analysis.

Disagreeing with political spin is reasonable. Replacing it with an unsupported trillion-dollar figure and claims of bad faith is not.

If there's a specific system design you think governments should commit to, the intellectually honest step is to specify its assumptions and cost that system, not assert that everyone else is hiding the answer.

To John Daysh
You wrote: "....detached from assumptions about demand growth, reliability standards, technology mix, financing and time horizon. That number does not exist in any serious modelling framework, and inventing one while accusing others of concealment isn't analysis." So what you're really saying is that the federal government, using taxpayer money, should be free to build a new electricity-production system in Australia for an unknown cost without any plan that outlines what reliability standards will be applied or or what the technology mix etc will be. Do you not understand how irresponsible such a process is for a government when the cost is almost certainly going to be hundreds of billions of dollars and in my view close to one trillion dollars? I say again: committing Australia to such a course of action is highly irresponsible. It is extremely poor governance,

Bernie,

No, that is not what I am saying. In fact, I've made a conscious effort not to say anything like it.

Governments do not build electricity systems "without a plan", and no one is proposing that reliability standards, technology mix or time horizons be left undefined. Those things are set through policy, regulation and planning processes, and then modelled through AEMO's ISP and related work. That is how system planning is actually done.

What I'm pushing back on is the idea that there exists, or should exist, a single, definitive "ultimate final cost of Net Zero" that can be published independently of those choices. Costs are produced after assumptions are specified, not before. Change the assumptions and the total changes. That isn't irresponsibility, it's how large infrastructure systems are planned.

Calling that complexity "unknown cost" or "no plan" is simply wrong. And repeatedly asserting trillion-dollar figures without a defined system, assumptions or source doesn't make the governance argument stronger, it replaces it with conjecture.

If you want to argue that a particular set of policy choices is irresponsible, then the argument has to be about those choices. It can't rest on inventing a single headline number and treating its absence as proof of negligence.

To John Daysh
So please show me the plan for Australia's future electricity generation and distribution network.
And it's the lack of a plan (or at least the lack of any publicly available plan) that is irresponsible, not the particular set of policy choices because I haven't seen any such policy choices and hence don't know if they're irresponsible or not.

There *is* a plan, Bernie. It's the one produced by the statutory market operator.

The national planning framework for generation, transmission, reliability standards and future system development is the Integrated System Plan, produced by AEMO. It sets out the scenarios, the projected mix, the required transmission build-out, the reliability settings and the system pathways.

It's publicly available and updated regularly.

That is the document governments use to assess system development. It is also where total system costs are modelled under different assumptions. GenCost is an input-cost report, the ISP is the system plan.

The absence of a single headline number doesn't mean the absence of a plan. It means the plan contains multiple scenarios with different trade-offs, because system cost is not a one-number question.

If you believe the ISP is inadequate or incorrect, that's a separate argument. But saying there is "no plan" when the national system plan is published, updated and widely cited simply doesn't reflect the reality of how energy planning is done in Australia.

To John Daysh:
"There *is* a plan, Bernie. It's the one produced by the statutory market operator. The national planning framework for generation, transmission, reliability standards and future system development is the Integrated System Plan, produced by AEMO. It sets out the scenarios, the projected mix, the required transmission build-out, the reliability settings and the system pathways."
So what is the cost of the plan, please, based presumably on what is considered to be the most reliable and cost effective scenario?

Bernie,

Again, there is no single cost of "the plan" because the Integrated System Plan does not nominate a single preferred system with a single reliability standard, technology mix and time horizon. It explicitly models multiple scenarios, each with its own total system cost, trade-offs and risks.

Asking for the cost "based on the most reliable and cost-effective scenario" assumes agreement on what level of reliability is required, what risks are acceptable, and over what timeframe costs should be assessed. Those choices are not technical outputs of the ISP, they are policy decisions that governments must make. Until they are fixed, there is no single number to quote.

The ISP does publish total system cost ranges for its scenarios. What it does not do, and cannot honestly do, is collapse them into one authoritative headline figure and present it as "the cost of Net Zero". That would require locking in assumptions that governments have deliberately kept open.

If your argument is that governments should explicitly choose one scenario, own its assumptions, and accept the political consequences of that choice, that's a political argument. But the absence of a single number is not evidence that there is no plan or that costs are being hidden.

To John Daysh
So there is a plan and there is an estimated costing. Why have you tried to deny for so long that a plan, regardless of its various shortcomings and uncertainties that you have referred to, did not exist?
AI says the cost of this plan is some $122 billion but it also says: "The $122 billion estimate covers utility-scale generation, storage, and transmission, but excludes consumer energy resources (like rooftop solar), distribution upgrades, and already committed projects." Some 4.2 million households have solar panels on their roof, costing roughly $21 billion. What does the plan say about expanding this number? My guesstimate is a doubling at least but with most of these houses also having batteries to feed back into the grid when needed, so another 4 million homes with $15,000 solar panels and batteries will add another $60 billion to the cost.
When you ask Google "how much money has been spent to date by governments around Australia on renewable energy projects?", the answer comes in at about $200 billion when you add up what's been spent to date. So the total is now approaching $400 billion. Do you not see why people like me are so concerned, when much of this expenditure will have to be spent again as solar panels and batteries need replacing every 10 to 20 years. No wonder electricity prices are going through the rood - Nero (Bowen) playing the violin while Rome (Australia) burns!

I haven't denied the existence of a plan at any point, Bernie.

In fact, I've repeatedly said the opposite: that Australia's system planning is done through AEMO's Integrated System Plan, and that GenCost feeds into it as an input-cost study. What I've pushed back on is the claim that there is no plan because there isn't a single headline number labelled "the cost of Net Zero".

Those are not the same thing.

The figures you're now assembling illustrate exactly why collapsing everything into one number is misleading. You're mixing ISP system-cost estimates, private household investment decisions, historical expenditure, replacement cycles, and speculative future uptake into a single running total, as if they were all the same category of cost borne by government.

They aren't.

Rooftop solar and household batteries are voluntary private investments made for private benefit, not line items in a government-built system. Counting them alongside transmission and utility-scale generation as if they were the same thing is a category error.

Likewise, adding past spending to future system scenarios doesn't tell you "the cost of the plan", it just accumulates unrelated expenditures across different policy regimes, time horizons and asset lives. By that logic, you'd have to add the replacement cost of every coal plant, gas turbine and transmission asset ever built as well.

None of this is an argument that costs are trivial or that electricity prices aren't a serious issue. It's an argument that meaningful discussion requires consistent boundaries. When you mix public and private spending, historical and future costs, and utility and consumer assets into one pot, the number will always look alarming, but it won't describe any real system anyone is proposing to build.

If the concern is affordability and governance, the productive way forward is to compare like with like: defined scenarios, defined assumptions, and defined scopes. Chaining together guesstimates across incompatible categories only creates heat, not clarity.

To John Daysh
I think you are being deliberately obtuse and misleading. Nowhere in my comments listing all the costs of our new electricity system in Australia did I say that the costs were to be borne by the taxpayer. I simply gave a summary of ALL the costs, private and public. Since you're not denying the magnitude of the cost total that I provided - about $400 billion - I'll accept that it's in the correct ballpark.

Bernie,

I'm explicitly NOT accepting that figure, and I've said why. Declining to endorse a number built by aggregating incompatible categories isn't conceding it's "in the right ballpark", it's refusing to validate a meaningless total.

What you've done is add together private household investments, public infrastructure, historical spending, future projections, asset replacements, and speculative uptake assumptions across different time horizons, then present the sum as if it described a single system cost. It doesn't. It describes an accounting collage.

Whether those costs are borne privately or publicly isn't the core issue. The issue is scope and consistency. Once you mix voluntary household purchases, government network investment, sunk costs, and hypothetical future replacements into one running total, the result will always look large, but it won't correspond to any plan anyone is proposing or any system that's being modelled.

I'm not being obtuse. I'm insisting on basic accounting discipline. If you want to discuss affordability seriously, the only defensible way to do it is to compare like with like: a defined system, over a defined period, under defined assumptions. Anything else produces a big number, not an answer.

To John Daysh
But you've previously told me that "a defined system, over a defined period, under defined assumptions" doesn't exist and hence can't be costed. I'll stick with my trillion dollar cost estimate until someone comes up with the details of your "defined system, over a defined period, under defined assumptions".

No, I haven't, Bernie.

//But you've previously told me that "a defined system, over a defined period, under defined assumptions" doesn't exist and hence can't be costed.//

What I *have* said, consistently, is that there is no single, pre-political, immutable system with a single cost that exists independently of assumptions. Defined systems do exist, and they are costed, but plural, not singular. That's exactly what the ISP contains: multiple defined systems, over defined periods, under defined assumptions, each with its own cost and trade-offs.

Saying "there is no single ultimate final cost" is not the same thing as saying "there is no defined system" or "no costing". You've repeatedly collapsed those into the same claim, despite being corrected.

If you choose to "stick with" a trillion-dollar figure until a single magic number appears, that's your prerogative. But that figure isn't derived from any defined system, assumptions, timeframe or model. It's a placeholder for dissatisfaction, not a cost estimate.

At this point the disagreement isn't about whether plans or costings exist. It's about whether you're willing to accept that serious infrastructure planning produces ranges across scenarios, not one number to end the argument.

Serious planning produces a plan, surely. There may have to be some preliminaries, but before the work starts there needs to be a plan, a management structure and a budget. A scenario is not a plan. And the envisaging of possible scenarios is an exercise in ..... envisaging scenarios. I had the same argument when the first AEMO Integrated System Plan was published. And the same confusion followed. I had the feeling I was on a different planet. And I still cannot see how construction could proceed from the present state of affairs. Of course that may be an unreasonable expectation, given that the whole project seems to depend on setting targets and attracting investors, rather than allocating funds and building a working clean power generation and distribution system that serves the needs of its customers.

TomBie,

This is where I think the confusion keeps arising. You're using "plan" to mean a single, fully specified engineering project with a fixed design, fixed budget, fixed timeline and a management structure ready to pour concrete. That's a reasonable expectation for a specific asset. It's not how national electricity systems are planned.

The Integrated System Plan is not a construction plan, and it was never intended to be. It's a system-level planning framework that answers a different question: given different policy settings, demand trajectories and reliability standards, what kinds of generation, storage and transmission are likely to be needed, and in what broad sequence? That's why it presents scenarios rather than a single blueprint.

Actual construction proceeds downstream of that framework. Individual projects, whether transmission lines, generators or storage, then have conventional project plans, budgets, approvals and management structures. Those don't appear at the ISP level because they can't, until governments, regulators and markets make specific choices.

In other words, scenarios aren't a failure to plan, they're the necessary precursor to committing capital in a system where demand, technology costs and policy settings are still moving. Expecting the ISP to look like a turnkey construction plan is like expecting a defence white paper to include the build schedule for every future ship and aircraft before cabinet has made the decisions.

You're right in that this approach is very different from a centrally funded, centrally built program. But that's a political choice about market structure, not evidence that planning is absent or incoherent.

To John Daysh
Thank you for your last comment. It explains quite well why the failure to produce a plan is producing a new electricity grid that is a dog's breakfast and hence why it's so expensive. As you say, a scenario is not a plan nor is a framework a plan and it is sheet stupidity to make investment decisions based upon a range of scenarios - a vague framework - that lack an overarching coherent plan.

No, it doesn't, Bernie.

//[Your last comment] explains quite well why the failure to produce a plan is producing a new electricity grid that is a dog's breakfast and hence why it's so expensive//

I have said the opposite, repeatedly. I have explained that there is a plan, that it is publicly available, and that it is deliberately structured around scenarios because that is how large, complex systems are planned under uncertainty. Recasting that as me "explaining the failure to produce a plan" is simply untrue.

What you keep doing is redefining "plan" to mean a single, fixed, centrally specified blueprint with a locked budget, and then declaring everything else to be "no plan". That's not an argument about evidence, it's an argument about semantics - and one that ignores how electricity systems, defence capability, water infrastructure and transport networks are actually planned in the real world.

You're free to argue that Australia should have adopted a fully centralised, command-and-build approach instead of a market-based one. That's a political position. But asserting that scenario-based system planning is "sheet stupidity" or evidence of absence of planning is not honest engagement with what I've said or with how these systems are governed.

If you want to criticise the outcomes, criticise them. But stop attributing positions to me that I haven't taken in order to do it.

So we've reached the point where two different kinds of "plans" have been defined and discussed, one that's represented by the AEMO ISP and the other that simple folk like me imagined was a normal plan, like the document that enables construction of something, like a shed or a home or a factory for making, say, lithium batteries.

It seems that the second kind was chosen for our new electricity supply by, well, somebody (anon). My questions are: How was the choice made? Did that process include consideration of whether it would work, something like a feasibility study I'm guessing? Were any working examples picked out for study? Has an ISP -style plan been successfully implemented in Australia? Was there any evidence from prior Australian planning practice that encouraged those decisions? Is the process that led to adopting the ISP available, in the public domain. And did the planners, or whatever they like to be called, consider that a fleet of clean electricity generators running on nuclear power could be connected, cheaply, to an existing transmission system and hence to existing customers. They might have seen the nuclear ban as an obstacle but then they might have seen the transmission cost savings as a favorable benefit to influence public opinion. Of course we know that none of the above came into any planning discussion. A decision had been made long before any other decisions (or whatever it is these strange planners make) were even thought about

Finally (?), First: Energy planning -- the big picture: I've already asked about the origins of Australia's ISP and the existence of evidence that might guide me, or others, on whether it will or can work. Nothing so far, but still worth stressing. Why not look at the countries considered most successful so far in their transitions. Which are they? How did they go about their planning? Might give us some clues. But to my mind the evidence is that no such question was ever asked and the plan we do have in Australia, the ISP, is the product of some kind of strange bureaucracy with no evidential basis.
Second, progress of the transition in Australia has made plain that the current plan should have been rejected on two simple predictable cost grounds. No.1, the cost of firming an intermittent solar/wind system is huge. No.2, Transmission costs for such a system, which prides itself on its "distributed energy" description, are extra huge, which must have been obvious to any engineer involved. The system was meant to replace a national network spanning thousands of kilometres, of mainly gigawatt-scale coal fired power stations. Each of those could in principle be replaced by nuclear energy. New transmission costs = zero. That's big savings, surely attractive to someone in authority, who could kick off an effort for Australia to remove its totally irrational ban on nuclear power. What actually happened? Total silence. It takes a long time to change cultures.

A few points, TomBie.

First, the ISP didn’t emerge from a vacuum. It builds on decades of NEM planning, regulator-set reliability standards and published cost inputs like GenCost. It’s consulted on, revised and stress-tested publicly. You can disagree with its conclusions, but calling it “a strange bureaucracy with no evidential basis” ignores how it was developed.

Second, firming and transmission costs were not a surprise discovered halfway through the transition. They are explicitly modelled. You can argue they are too high relative to alternatives, but they weren’t invisible to planners.

Third, the idea that replacing coal with nuclear means “new transmission cost = zero” isn’t realistic. Not all coal sites are suitable for nuclear; licensing, cooling, seismic, financing and timelines matter; and grid reinforcement would still be required. The nuclear ban is a legislative decision, not something planners quietly forgot about.

The disagreement here isn’t whether planning exists. It’s whether Australia should have pursued a centrally directed nuclear pathway instead of a market-based renewables pathway. That’s a legitimate policy debate. But it’s not evidence that planning lacked coherence or evidence.