Richard Willoughby
Summary
This article analyses one year of wind and utility solar generation data from Australia’s National Electricity Market (NEM). Using generation and curtailment data from OpenNEM, it estimates the storage requirements needed to firm wind and utility solar generation, examines the cost trade-offs between battery storage and generation overbuild, and extends the assessment to pumped hydro, household battery subsidies, and the economics of grid-scale storage. The author concludes that:
- Wind and solar show some complementary behaviour (“synergy”) when combined.
- A 16-hour battery appears to provide the lowest overall system cost based on the stated assumptions including current ratio of wind to utility solar.
- Supplying peak NEM demand using only wind, solar and batteries would require substantial additional generation and storage capacity.
- Based on the stated cost assumptions, the resulting levelized cost of electricity (LCOE) would be twenty times higher than conventional lignite fired generation costs achieved in 2003.
- Recent market developments reinforce the author’s view that large-scale wind, solar, batteries and pumped hydro do not currently offer favourable economies of scale in the NEM.
- The Discussion argues that Snowy Hydro’s useful storage volume is constrained by its relatively low generating capacity, while home battery subsidies and expanded small-scale solar eligibility are shifting value away from grid-scale batteries and pumped hydro.
- The author concludes that a fully renewable NEM remains economically impractical under current conditions. Rooftop solar and behind-the-meter batteries are more economically viable than centralised “renewables” projects for the emerging de-industrialised economy.
Wind & Solar Generation in the NEM
Australia’s National Electricity Market (NEM) supplies 86% of the Australian population through a single interconnected grid. The registered wind capacity is 13.46GW and the registered utility solar capacity is 10.71GW.
In the year to 1st August 2026, wind generated 36.6TWh and utility solar produced 19.8TWh. The recorded daily generation is displayed in Chart 1.
The key observations from Chart 1 are that the wind varies considerably day-to-day with multiple consecutive days of low output while the utility solar has smaller day-to-day variation but the annual variation is clearly evident.
OpenNEM also records the estimated curtailment of both wind and utility solar making it possible to assess the true potential of these two power sources. Chart 2 shows the cumulative generation less the daily average over a year.
Chart 2 actually shows what the installed capacity is capable of if there was sufficient storage to absorb all excess and then release it as required. The average daily output for wind would be 111GWh and would require a battery capacity of 1700GWh. The existing utility solar would be able to supply daily average of 65GWh when supported by a 2600GWh battery.
Trade-off of Battery Capacity and Generation Overbuild
At present battery costs, it is more economic to overbuild the generation to reduce the size of the battery. Chart 3 shows the charge state of batteries with 24-hour storage capacity working independently for wind and solar.
In this situation, the existing wind could support an average daily demand of 48GWh while utility solar could support 34GWh per day.
Chart 4 shows the result of combining the sources of generation while still using a 24-hour duration battery.
The combined daily demand that can be served is 99GWh. So the two sources exhibit synergy; enabling the combined output to be 20% higher than the sum of the individual outputs.
The Minimum System Cost Battery
The current costs for the key components of a battery firmed wind and solar grid are:
- Wind Turbines – AUD3.5bn/GW
- Utility Solar – AUD1.5bn/GW
- Battery AUD0.5bn/GWh
- Transmission Lines – 30% additional capital for generators
Using these current estimates and the demand pattern with the existing wind and solar generation, it was determined that a 16-hour battery gives the lowest overall system cost. Chart 5 indicates that such a system could meet a daily demand of 93GWh.
100% Battery Firmed Wind and Solar System
The highest daily demand was 728GWh so the existing wind and solar generation would need to be scaled 7.8 times to get from 93GWh per day to 728GWh. This gives a system requirement of:
- Wind Generator Capacity – 105GW
- Utility Solar Capacity – 84GW
- Battery Capacity – 485GWh
Levelised Cost of Energy from Wind and Utility Solar
Using the system cost components, the total system cost comes to AUD958bn. Applying a cost of capital of 7%pa and O&M cost of 2% of capital per year over a 20 year life requires an annual income of AUD110bn. Considering the annual demand totalled 220TWh, the average unit cost would be AUD500/MWh.
This cost estimate is based on further simplifying, but optimistic, assumptions such as:
- 100% round trip efficiency for the battery.
- Battery systems can support most FCAS requirements while any inertia requirements are covered by the 30% transmission margin.
- Market management costs are in addition to the LCOE calculated here.
100% Battery Firmed Utility Solar
The existing utility solar can produce 34GWh per day with a 24-hour battery so it would need to be scaled 21 times to meet the peak daily demand with a purely solar/battery system. This reduces the capital cost to AUD920bn and unit cost to AUD478/MWh.
Discussion
The mechanism that encouraged weather dependent generators onto the grid was legislated in 2000 in Australia. The price impact was noticeable after 2003 when lignite fired generation was selling for AUD23/MWh. Lignite generation now sells for AUD85/MWh because it is being cycled rather than operating at full capacity continuously but remains one of the ESSENTIAL sources of generation so can charge high prices when wind and solar are low.
Despite there being synergy with wind and solar generation, the high cost of wind generators in Australia makes them uneconomic against utility solar. And utility solar is uneconomic against rooftop solar because rooftops have dedicated demand. The low financial close on projects under the Capacity Investment Scheme indicates that investors are faced with actual costs exceeding estimates thereby making projects uneconomic despite the government guaranteed return on estimated cost.
Snowy Hydro pumped storage provides a storage capacity of 350GWh but its generating capacity is limited to only 2GW. On a 16-hour basis, it only offers 32GWh. So at the estimated cost of AUD42bn it works out at AUD1.3bn/GWh – more than double the current battery cost but it should outlast some five sets of battery using current technology.
It is noteworthy that in Q2 2026, the 240MW Shoalhaven pumped hydro energy storage (PHES) earnt just $1.3M for the quarter. There is now a clear decline in income for PHES in the NEM since the introduction of the Cheaper Home Battery Scheme (CHBS) from July 2025 per Image 1 extracted from the AEMO report. The CHBS scheme provides a 30% subsidy for battery purchases resulting in 11GWh being installed in the past twelve months; much faster than grid scale projects and households now own most of the battery capacity on the grid with three more years for the program to run.
A further recent observation is that the arbitrage on grid scale batteries has collapsed from $360/MWh a year ago to $60/MWh in the latest quarter. So the CHBS has now made grid scale battery installations uneconomic.
As further evidence of the economic reality of negative benefit of scale for wind, solar and batteries, the Federal Government has made a 10-fold increase in the small-scale solar limit; taking it to 1MW from the existing 100kW. That means shopping centres, hospitals, schools, etc can now gain subsidies for large solar/battery installations.
Conclusions
Building a 100% “renewables” grid for the Australian NEM remains firmly in the realm of fantasy for the present time. Achieving the first 50% has caused a 4-fold increase in base load cost and the next 50% will cause another 4-fold increase.
Snowy Hydro pumped storage will only enable one more coal fired power station to be retired providing wholesale demand continues to slowly decline or is static. Its power rating is inadequate to fully utilise its storage capacity. If economics prevailed over insanity, the Snowy Hydro pumped storage project would be stopped. Its earning potential at the present time is just $40M per year or 1/1000th of its capital cost. The potential income will not even cover its operating and maintenance cost. It is a comprehensive waste of money; albeit the Labor Party receive generous donations from the unions to keep the money pit producing.
Wind, solar and batteries do not offer any economy of scale. That had previously been observed with the rapid uptake of rooftop solar in Australia and the steady decline in the wholesale market. Home batteries are now making the same point as the grid scale batteries and pumped hydro lose their market.
The de-industrialisation of the Australian economy continues per Image 2 with quarterly volume down 0.5% year-on-year as rooftops exceed any growth.
The Author
Richard Willoughby is a retired electrical engineer having worked in the Australian mining and mineral processing industry for 30 years with roles in large scale operations, corporate R&D and mine development. In the early 1990s he represented large customers on the first market system committee for the Australian National Grid. A further ten years was spent in the global insurance industry as an engineering risk consultant where he developed an enduring interest in natural catastrophes and changing climate.
Good post with much granular detail—so will surely go right over most Aussie politician heads.
The two fundamental problems with grid wind and solar can be simply stated such that even politicians might understand. Renewables are intermittent, while the grid must run continuously. So renewables must have some sort of grid backup.
Australia is not the only country suffering deindustrialization as a result of these hard grid realities. UK and Germany are also suffering similar consequences.
As the article says, “de-industrialised economy”. One of the costs that is not immediately apparent is what the cost of this will do long term. A recent article I read discussed the organizational knowledge being lost. It was mainly addressing electrical substation transformers but applies to a lot different disciplines. The skill in creating the transformer steel, in winding the coils, and designing the apparatus is quickly disappearing in the United States. Colleges and universities do not have the expertise to train new people. It is a national security issue. It is apparent in the recent collapse of a New York add-on to a high-rise building. I know of two businesses who can’t find highly trained machinists. They must train people themselves.
Lots of unknowns follow on these energy policies that have never been considered.
” twenty times higher than conventional lignite fired generation
costs achieved in 2003 [AUD23/MWh]”
There is a reason for the mention of 2003 here. First, it is just a few years after the operaations were privatised. But second, since 2003 prices of most fossil fuels here have risen greatly, due to he expansion of export facilities. So, as you go on to say, lignite based power is now AUD85/MWh, and is still a lot less than the black coal generation of NSW and Qld.
So the hypothetical cost in this calculation is not so much greater than current FF cots.
Dan Andrews tripling of royalties to the cost of lignite coal in Vic in 2016/17 would be most of that price rise one might think?
Dan was a world expert in pissing in the soup, wasn’t he?
Brown coal in Victoria was never exported, so there is your first misdirection.
Let’s pray that they keep Yallorne power station open way past 2028 !
Let’s pray that they keep Eraring power station open way past 2029 !
Another factor causing the cost of coal fired generation to increased in NSW and Qld is because it constantly has to ramp up and down, when the are most economically efficient when run 80%+ of capacity.
They now have to cover all their maintenance costs while running at partial variable output, (which increases their maintenance costs).. so of course the price goes up.
Each eastern state should be building a couple of new, high efficiency COAL fired power stations right now…
… before it is too late. !
There is no export market for lignite from Victoria. The only increase in fuel cost is due to the 3-fold increase in State royalties to tax the coal fired generators more.
Even after tripling, royalties are tiny. About 0.23$/GJ, or about 0.8$/MWh. About 1% of electricity cost.
Brown coal is not exported, but the competitive black coal is, hugely, and that flows through to the cost of brown.
“and that flows through to the cost of brown.
NO, it doesn’t !
I was only countering your point on coal prices increasing. The only increase for Lignite is State royalties.
Your MWh is not electrical megawatthours either. You have to allow for thermal and mechanical efficiency.
With regard the $85/MWh, that was the average for 2025 year. Q4 2025, the lignite generation was priced at $47/MWh. In the 4th quarter they had to sharpen their pencil to keep above their 60% lower limit when to push more grid wind and solar out as rooftops gobble up demand.
I expect the lignite generators could still make money at $30/MWh if run flat out.
The AEMO charge now works out at $3.83/MWh on top of the wholesale price. Almost $1bn annual cost; far removed from what was contemplated by all eight of us when we first met in Canberra in 1993 to begin the design of the National Grid market system.
The AI industry is already locking up land in Latrobe Valley. They will await a more amenable State Government to access the fuel.
AUD85/MWh is still a factor of 6 less than the 500AUD/MWh for the all renewables approach.
Well, the 500 is RickWill imagination. And the 85 is only available in Victoria. Much higher elsewhere.
And the 85 is only available in Victoria. Much higher elsewhere.
You have no idea Nick. My whole incentive to push for the National Grid back in the early 1990s was to get access to the Latrobe Valley generation for our operations in NSW and SA.
I was offered Lignite generation at $40/MWh in 1989 when we were paying $60/MWh for electricity in NSW. We payed for an interconnector to Wentworth but is was built by Elcom. When SECV and I floated the idea of wheeling Latrobe Valley generation through NSW to our operations, they said the wheeling charge would be $20/MWh. We had paid for the power lines so they were simply protecting their monopoly.
Look at the productivity boom in Australia that followed removal of State monopoly power on electricity supply.
Now we have an electricity grid focused on fixing the weather that is far from broken.
When you have your house running off-grid, send me your invoices for the capital and operating costs and we will then know you have some real knowledge.
“Look at the productivity boom in Australia”
But you’re telling us that industry collapsed because of high electricity prices???
“You have no idea Nick”
I can’t see what you said that is not in line with what I said. The price of brown coal elec is not typical, and is much more elsewhere. OK, you can do what you describe to get around it. But most can’t.
Again – no idea Nick.
The productivity grew under Howard tanked under Rudd-Gillard-Rudd and was positive again under Abbott.
Estimates of Industry Multifactor Productivity, 2024-25 financial year | Australian Bureau of Statistics
And productivity “growth” is now negative.
Frustratingly, ideology and rationality cannot function in the same mind space at the same time.
Even in Nick’s big brain.
Lignite is impossible to safely transport any distance , it catches fire spontaneously.
It remains a local resource, not directly connected to international prices
Lignite’s price is controlled by what LOCAL buyers are prepared to pay
There is no meaningful price of lignite; the relevant number is that price at which using it they can generate electricity, and thus compete with the black coal generators (and gas, solar etc).
Blasphemy! Heresy!
(Tickets for the 3pm stoning available from Labor headquarters)
Ya know, when GHE fails all of this goes with.
There is a bit of both. The economy destroying cost of “renewables” has created momentum to really assess if the medicine is not worse than the illness.at
That is quite different to recognising CO2 is a trivial player in the atmosphere.
Bill Gates, for example, still thinks CO2 is a problem but not quite as bad as having no energy to run Microsoft’s AI.
Bill Gates said over 10 years ago that it is impossible to run a modern industrial society on renewables.
But while the Biden crew were running the joint, there were still untold $billions of taxpayers $$$s being handed out for “renewables” grifters.
So Bill being Bill, he wasn’t about to abandon the hand that fed the grift.
I might have missed a point or two in the article, but can you show where the replacement cost of the batteries and W&S are included in the cost/MWHr?
Surely a battery which is heading toward the $1T mark and a lifespan of say 10 years, would add $100B to the costs each year. This does not appear to be included in your costings.
I chose a 20 year life and allowed a 2% O&M charge to avoid major replacement cycle. I doubt any will ever be replaced but after 20 years you start again if the economy still functions.
As you can see from Chart 5, the battery really only works hard for a few days a year. There will be a little more day-to-day swings than is possible to show using daily data but it is the big dips that matter from the perspective of battery life.
I have large format Winston cells on my off-grid system and they are still doing well in their 15th year. Solar panels are in their 17th year and last year was higher than the three previous years due to less curtailment after my grid battery was installed.
My battery is 48-hour duration but I have no diversity of supply while the grid solar clearly benefits from supply diversity and being positioned away from the coast.
Very nice, wind and solar can’t support the grid, everybody knows that. Stop wasting money on them.
“Workers on Australia’s biggest renewable-energy project, Snowy Hydro 2.0, will receive upfront pay rises of at least AUD50,000 under a new deal that will see their annual wage soar to AUD300,000 pa” (The Australian 9 Sep 2025).
A sort of circular economy or crony socialism.
The Australian government already subsidizing the Whyalla steelworks and the Mount Isa copper smelter is now about to support Australia’s biggest aluminium smelter at Tomago, all unsustainable.
Half of the aluminium smelters now gone. All road vehicle manufacturing gone.
Shipyards are government funded boondoggles.
There is view that AI data centres will eventually increase wholesale demand but the big AI firms are already courting State Governments for direct access to fuel. Their overall cost structure in the USA is 40% energy so low cost electricity is a serious competitive edge.
SRES limit increase to 1MW will see a big boost in rooftop solar.
Australia did not change much in the last 100 years. Back then, you could be jailed for working more than 8 hours a day.
What this analysis means is that wind/solar/battery systems are net energy consumers. The additional cost of their operations is actually the cost of conventional energy used to construct and operate them. If they were never created, less fossil fuels would have been used for the same delivered energy.
Back in 2010 when I purchased 3kW of solar panels for $9,000, I could have purchased 90 tonnes of anthracite and sent it off to a power station to supply me with electricity for 50 years – lignite costs far less than anthracite. They already had the generating asset and 90t was a few minutes of operation. All the transmission lines also existed.
My purchase was incentivised by government sanctioned theft from electricity consumers. They paid $3,000 up front and $660/MWh for the energy I exported. I made sure I exported all I could – typically 3MWh per year. That program ran for 14 years and meant I had zero cost electricity with export income more than paying for imported electricity.
Panels are lower cost now and batteries are lower cost as well but it requires a lot of both to run a house located at 37S through all seasons and weather conditions.
If solar panels and batteries lasted for a couple of hundred years, they would be able to return the energy used to make, ship and mount them.
One further point to add to the problems of wind and solar is the lack of severe thunderstorms in recent years which will return. Then, large scale destruction will occur because no wind turbine can be designed against for two types of wind turbulence.
The extreme downbursts under active thunderstorm cells which drive all blades down regardless of being feathered and these can exceed 200 k.p.h. The rapid wind change when the passage of storms are associated with cold fronts. In that instance the wind veers 180 degrees suddenly which no steering mechanism can overcome and then there occurs the resultant huge stress on gear box and bearings.
The gear box and bearing failures are quite high world wide and full data disclosure is never published for obvious reasons, as are damage reports from hail and lightning.
Victoria’s last major blackout was due to a weather front that first caused wind turbines to shut down on high wind, then took out major transmission towers then caused a dramatic reduction in solar output. All that over a matter of minutes. The lignite fired generators could not cope so some 100,000 consumers were shut down to unload the generators.
More grid batteries have been installed since and they are not as prone to wind conditions and most have 4-hour rating with sub-second response capability. Batteries actually make the grid more resilient.
I didn’t even mention batteries because they do not generate power. They store chemical energy which can release electrical energy which I am sure you know. The point I was making was that wind turbines and solar panels will be destroyed by severe thunderstorms when they occur again. That power generation won’t be lost for four hours or even four days but years or may not get rebuilt again.
The Victorian weather you speak of was NOT caused by severe thunderstorms but a front as you say which would have contained normal sized thunderstorms with tops around maybe 30,000/35,000′. That is not the type I am talking about, but lines of really big storms in summer with the right meteorological conditions which we have not seen for nearly 40 years. Most people have not experienced them that is why they are not considered when talking about power or the building of intermittent generators sitting out in the open in exposed areas.
“…existing wind could support an average daily demand of…”
Meeting or not meeting average daily demand is not the issue for electricity grids. Similarly levelized cost of electricity at the point of generation is also not the issue. Instead, the issue is that the grid, whether a national grid or an assembly of local grids, must supply of enough electricity to meet demand, whatever that is, minute by minute or even second by second. The levelized cost and average supply arguments for grids are akin to arguing that if each home had an oxygen source, one could, for example, double the amount of oxygen in the homes air and breathe only for 12 hours per day thereby providing the same average amount of oxygen to each person for less work. Humans don’t work that way. Neither do electricity grids.
You have missed the point of the whole article. The levelised cost calculated in the article is determined on the basis of supplying electricity continuously including the highest demand day.
The $500/MWh is an order of magnitude higher than what is often quoted for the levelised cost of wind generation because the usual method of calculation ignores intermittency.
Battery storage for the electricity generated intermittently by weather dependent wind turbines and solar panels DOES NOT RUN THE ECONOMIES OF THE WORLD.
Net Zero enthusiast are OBLIVIOUS to REALITY that the economy runs on transportation fuels.
· Jet fuel for the military and international airports,
· Diesel fuel for trucks and construction equipment,
· Different grades of gasoline for vehicles, and
· Bunker fuel for all the ships arriving and departing from Ports.
In addition, raw crude oil, once processed through refineries provides the basis of more than 6,000 products in our materialistic society that are made from those oil derivatives.
Wind turbines and solar panels can ONLY generate electricity but CANNOT make any of the products or transportation fuels MADE FROM oil.