We keep getting told that wind and solar renewables are cheap, yet our bills keep going up. So, what’s going on?
From the EIGEN VALUES Substack
Last week, I decided to write a Twitter/X post to summarise how much we are paying for renewables. It got far more traction than I anticipated, so I thought it would be helpful to convert it and extend it a little to make a bonus article on Substack that can act as a succinct response to all those who still insist on claiming renewables are cheap.
In the UK, renewables are subsidised by three different schemes. Feed-in-Tariffs (FiTs) fund mostly solar power. The latest report for 2022-23 shows the scheme cost over £1.7bn and average total payment was ~£193/MWh, about 3X the current cost of gas-fired power at around £65/MWh (see Figure A).
Contracts for Difference (CfDs) fund a range of technologies, but most of the subsidy goes to offshore wind. Latest data from the LCCC shows the subsidy per MWh fell dramatically during the energy crisis, but is now back at £95/MWh for offshore wind, £73/MWh onshore and £60/MWh for solar. April 2024 was a record month for overall subsidies with £268m paid out with average strike prices at £146/MWh for offshore wind, £113/MWh for onshore and £110/MWh for solar power (See Figure B).
The CfD subsidy for burning trees in biomass plants rose from about £7/MWh in March to nearly £60/MWh in April. This encouraged more biomass generation and the total subsidy paid for biomass jumped from £2m in March to over £34m in April 2024. The total CfD subsidy paid for the last 12 months is over £2bn, and the trend is clearly upwards.
By far the biggest subsidy scheme is Renewables Obligations Certificates, costing over £7bn per year. This scheme awards certificates for each MWh generated, a different number depending upon technology. April reference prices for intermittent renewables have been around £53/MWh, meaning the average price paid for offshore wind under the ROC scheme has been £176/MWh, onshore £118/MWh and solar £146/MWh including the value of the certificates (See Figure C).
Future renewables are also going to be more expensive than current market rates. The Government’s announcements of prices for their Allocation Rounds are often quoted in 2012 money. In 2024 money, Allocation Round 6 (AR6) is offering £102/MWh for fixed offshore wind, £246/MWh for floating offshore wind, £89/MWh for onshore and £85/MWh for solar power, all far higher than current market prices (see Figure D).
It should be noted that FiTs, ROCs and CfDs are all index-linked, so prices will continue to rise with inflation. It is clear our bills are going to continue to rise for the foreseeable future as cheap gas is forced out in favour of expensive renewables.
In addition, we pay extra for balancing the grid when the wind is not blowing (or blowing too hard) and the sun is not shining. In the year ending March 2024, these balancing services cost £2.46bn. Most of these costs should be attributed as a cost of intermittent renewables. That’s a total of over £12bn being paid to or because of renewables each year.
More costs are coming down the line as the National Grid ESO has announced £54bn of spending on the electricity network infrastructure up to 2030 and a further £58bn in the 2030-2035 period, a total of £112bn, or over £10bn per year for more than a decade.
I hope it is clear to all now that renewables are not cheap and are never going to be.
Eigen Values is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscri
Eigen Values is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
I’ve worked in distribution ops and transmission ops. I recently applied to a job in solar field O&M ops. I think it is lower stress. It’s not NERC ops so I really don’t understand their operating mentality. I kind want to get an inside view on this stuff. Also, I would like to ride the government subsidy chain for once. Pay me not for my value, but because you have no shareholder responsibilities.
But, But, But, the fuel source is freeeee…
It’s not about the fuel source dummy, it’s about the overall cost per MWh generated AND how much your government is willing to pay them to produce nothing.
Nothing’s cheaper than Coal or Natural Gas…and Gas is Natural
Wind is at least 4 times more costly than FF generation given longevity of facilities and requires 2.25 times nameplate capacity to equal FF generation capacity factor.
Wind is at least 3 times more costly than Nuclear Gen. given longevity of facilities and
requires 3 times nameplate capacity to equal Nuclear generation capacity factor.
FF generation lasts 80-100 years with regular maintenance before needing replacement.
Nuclear lasts mat least 60 years with regular maintenance before needing replacement.
Wind lasts maybe 20 (likely less) even with maintenance.
Solar lasts 10+15 years or more likely from Hail Storm to Hail Storm.
Both Wind and Solar produce toxic waste or plastic waste that can’t be recycled and must go to landfill at end of life
Wind will be replaced 3-5 times respectively over the lifetime of a single Nuclear generator or FF generator, each time at progressively higher costs
Solar will be replaced at least 4-6 times and likely more due to storm damage … At progressively higher costs
Always remind them all fossil fuels are free, however, just like ruinables, the cost money to extract.
Well, free to someone, sort of. I can assure you if there was oil or gas under my property, it would not be free to whomever would want to extract it.
And all produced “naturally” by Gaia!
AND much more land area taken… along with commensurate killing of endangered species of plants, birds, bats,….
Well said
I like your comment about the replacement cycle. I bet, that is never factored when they say ruinables are nine times cheaper than gas.
I once did the ‘back of a fag packet’ calculation on installing onshore wind farms. I’d estimated that, with planning permission delays, and build times, it would take around 20 years to build enough onshore farms to replace gas turbine delivery.
And I reckon, by the time they’re close to installing the last wind farm, the first batches will be in need of replacement.
I hope it is clear to all now that renewables are not cheap and are never going to be.
Well it’s really just a case of suppliers getting their pants down around their ankles-
Ford Pleads To Suppliers: ‘Help Us Make EVs Affordable’ (msn.com)
A bit of forced labour might help too rounding up those border crossing folk and putting them to work in the dark Satanic mills for their bug gruel and sustainable tents. Can’t have the masses idle and up to mischief in Green Utopia with the next 5 Year Plan and Great Leap Forward.
Ford should get back to building vehicles their customers want to buy, instead of closing production lines and behaving like communists.
Possibly they are closing to show the government the consequences of the EV legislation. If so, the gov’t isn’t catching on as quickly as the public.
The coast of renewables is flattered and disguised by failing to include the cost of the Energy Storage System (ESS) necessary to ensure 365/24/7 demand satisfaction.
At present, fossil fuels are the ESS used.
Eliminating fossil fuels requires an alternate ESS. This system will need to be sufficient to supply up to 45 days of demand.
So, if renewables are as cheap as its proponents claim, and data centres are huge consumers of electricity with a relatively steady and predictable demand, it would make economic sense for a data centre operator to construct its own independent renewable energy generation and storage system.
So, consider a data centre with demand 28.5 MW, or annually 250GWh in Ireland. Irish wind capacity factors are generally quite high at about 29%. However, in 2009, the average wind capacity factor for the year was only 20%, therefore this lower value must be used for long term planning..
At a capacity factor of 20%, to generate 250GWh of electricity over the year with wind turbines requires a minimum 142.5 MW install capacity of Wind Turbines before considering the inevitable inefficiencies incurred in charging and discharging the ESS. Lets assume an additional 28.5 nominal capacity would suffice, so total wing generation capacity of about 170 MW would be required.
ESS capacity required would be 40 days (being optimistic) x 24 hours x 28.5 MW = 27,000 MWh.
Costs:
Data centre with 28.5 MW demand @ur momisugly Euro11 million per MW demand = 313.5 million
170 MW nameplate capacity wind turbines @ur momisugly 2 million/MW = 340 Million
27 GWh Battery ESS @ur momisugly Euro 320/kwh (highly “optimistic”) = 8.640 Billion
Land area required for batteries @ur momisugly 40m3/MWh = 1.1 sq km, plus area required for switchyard and operations and maintenance facilities.
Conclusion:
the cost of providing an independent renewable power generation and storage system would likely cost almost 30 times the cost of the data centre itself.
The myth of “cheaper” renewables can only be sustained by fraudulent accounting.
Mike, I agree with your conclusion
but
A – using a single capacity factor from 2009 sounds like a cherry-pick; you should be able to use the average CF of at least 10 years.
B – using a capacity factor of 20% seems low;
(Average UK national capacity factor = 31.8 %
https://iea-wind.org/wp-content/uploads/2023/10/UK-2022.pdf )
C – If this 20% is a worst-case scenario over the last 15 years, then say so.
I have been collecting wind statistics over at least ten years – 20-23% is about average for capacity factors.
The IEAE statistics are not data, they are models. Based I believe on wind speed and take no account of turbine downtime…
A look at the actual performance of wind farms on CFDs
Note that colours move through the spectrum reflecting age of the wind farm. Some have only just taken up their CFDs although they may have been operating on a market price basis (much higher revenue) during the energy crisis. Note that Moray East has a very low capacity factor because it spends so much time collecting curtailment payments instead. This is going to be a BIG factor in future as capacity is expanded, with the contract terms being a major influence on which wind farms are first in line. Older wind farms get their subsidies either all the time or most of it (only if prices go negative for more than 6 continuous hours do some of them lose their subsidy for the duration). Newer ones lose their subsidies in any hour with negative prices, so they are first in line to curtail.
You can also look at Seagreen wind farm, which is still not on a CFD, so is entirely market price based. It is frequently curtailed as offering a better income than producing.
You can also see the impact of maintenance on output, with capacity factors dropping sharply for the duration when major maintenance is undertaken, perhaps includng the offshore substation.
The chart defines capacity as the maximum daily average output achieved in the life of the wind farm, which is perhaps a more realistic definition than the nominal capacity that is usually quoted.
Worth noting too that 2021 was a particularly challenging year for wind, with an extended period of low winds that would have required extensive backup.
I should also have noted that these windfarms are offshore ones, which tend to have much higher capacity factors than onshore ones. I’ll post a similar chart for onshore.
Here are the onshore wind CFD capacity factors. Complete shutdowns figure more often than might be expected
A – Absolutely NOT cherry picking. 2009 had the lowest annual capacity in Ireland since they started recording renewable performance. Since the object was to ensure 365/24/7 demand satisfaction, it is essential to use the lowest actual capacity factor to determine the required generation capacity. Use of averages brings you up against the dirty secret of averages, half of them are too high or too low, whichever is worse for you. Using averages without taking into account variation from the mean always results in inaccurate estimates.
B – Yes is it low. It is the lowest annual wind capacity factor experienced in Ireland.
C – I believe my statement clearly implied it’s the worst case scenario: “Irish wind capacity factors are generally quite high at about 29%. However, in 2009, the average wind capacity factor for the year was only 20%,” If the average is about 29%, clearly a value of 20% is a significant outlier. Boo, I wrote lower rather than lowest. Grammatically correct, 20% is lower that 29%, but i didn’t take the extra step of saying it was the lowest because I thought it was blindingly obvious it must the the lowest.
It can be worse than that if you are hoping to rely on storage, because you can get another below average year following on or preceding (or both!), so your storage based on just the worst year gets emptied. Storage round trips and protracted store times will consume energy too.
True enough. That would be part of the risk assessment process when determining the base capacity factor to use..
When I ran my simulation including 2009, early 2010 had winds strong enough to replenish the storage system by the end of April. If the low winds had continued for another couple of months however………………..
NB : I use a “brute force and ignorance” approach with BM Reports + ESO (30-minute) data.
The following estimates should be taken with a very large pinch of salt !
For the GB electricity grid I get “annualised” capacity factors in the 30-35% range, but note that the daily accumulator numbers can swing wildly from “less than 5%” to “more than 70%”.
Note also that I use the DUKES “ET 6.1” numbers for both Wind and Solar capacities, which are currently only available up to Q4-2023, combined with “The island of Great Britain = England + Scotland + Wales”.
“45 days”!
Some time ago, I did a calculation based on the opening of a new battery farm, near Hull. Apparently, that farm, could provide 1% of total UK demand for four minutes. The size of the farm was equivalent to two football pitches.
I’d worked out, that to provide a 24-hour supply, would require an area equivalent to the size of Greater Manchester.
So, that would be equivalent to 45 cities! 😕 Ludicrous!
Current battery farms in S Australia, US and Japan are about 40sq,m per MWh storage, plus switchyard and O&M facilities.
Scale that up:
GWh per 1sq km = 1,000,000 sq/m/40 sq.m/MWh
= 25,000 MWh/ sq.km
= 25 GWh/sq.km
Its much worse than this. This is adding up all the operating subsidies. But there is another very significant cost which is left out, the capital and operating cost of the gas installations needed to run what is in fact a hybrid gas + wind&solar system.
You have to have total gas capacity equal to the wind + solar supply you are counting on. The reason is the usual summer and winter calms. The winter ones happen when there is no solar. So at 5pm on a January weekday you have to be able to meet peak demand solely from gas.
It doesn’t matter how much you pay the wind and solar operators, for at least a week every winter you are going to get nothing from them, and if you want to keep the country going, you have to get it from gas. The same thing happens most summers, except you get more solar.
The argument to be made by renewable enthusiasts is not that wind and solar are cheaper than gas. They have to show that adding wind and solar to an existing gas network is cost effective because it lowers fuel costs. Because that is actually what we are doing. We are not powering x thousand homes from wind. We are powering them from gas supplemented by wind.
I have never seen a study showing this and don’t believe there could be one.
There are also other costs that are not taken into account by unreliables supporters. A major one being how they make life difficult for the grid operators. The shift to unreliables brings an end to highly controllable and dispatchable power generation and makes balancing the system increasingly complex whilst at the same time electrification of transport, heating and industrial processes, data centres etc means more and more electricity is being consumed.
Mitigating these developments and preventing grid collapse requires significant investment in back up capacity,as you say, but also in firming the grid and in storage solutions. Promoters of unreliables disregard all these costs when they claim unreliables are cheaper.
In other words, we are saying the same thing:
If the cost of renewable does not include the cost of the Energy Storage System they depend on for 365/24/7 demand fulfillment, whether gas, fossil fuels, batteries, pumped storage, etc., then the cost of renewable is not being honestly calculated.
Here’s what retailers are paying wind farms for their output
On average, it’s close to £150/MWh now. The line for Day Ahead Prices largely reflects the costs of the anticipated marginal sources of power averaged over the month. Much of the time this will be CCGT generation, although at times of higher renewables output imports (Coal/co-firing via BritNed, French nuclear, Norwegian hydro) will get backed out. The variety of biomass subsidies also result in reduced output at lower prices. At times of system stress prices cease to reflect cost, and instead reflect the prices at which demand (mainly industrial) backs off. Prices also cease to reflect cost when the renewables surplus becomes so large that it must be curtailed to maintain a stable system: then prices start to reflect how much wind subsidy must be eliminated to persuade wind farms to curtail. This is quite a different case from the frequent curtailment in Scotland because there is insufficient grid capacity to handle the output and transmit it Southwards – so mainly CCGT backup replaces the curtailed wind.
A look at last December at hourly resolution helps to identify what is happening. First, a more traditional presentation, with exports and pumping showing as negative volumes, so that all sources generation and imports can exceed actual use demand for power. The stack is designed to make it easier to see variations in baseload generation from nuclear, biomass and coal, and the pattern of interconnector use across the day, with the balancing role of CCGT against renewables and demand also easier to see. The marginal generators – OCGT, pumped storage, hydro are at the top of the stack so their role in filling demand peaks is clearer.
There is clearly colder weather at the start of the month with little wind, but it becomes windier later, and demand falls into the holiday period. Click on the chart to see a larger, clearer presentation.
Here is the same data as the previous chart, but simply with each hour sorted in declining price order. This helps to make clear the varying price response of generation and the impact of higher and lower demand. Again, click on the chart to see a larger, clearer presentation.
Do you believe it is a good pricing system?
This is not going to go down well.
Ausgrid is going to charge solar panels users.
They’re introducing a fee for users to export excess power back into the grid.
It will cost customers hundreds of dollars a year.
https://x.com/BenFordhamLive/status/1790891339317735816
It will incentivise households with rooftop solar to install batteries. That leaves those without rooftop paying more because the growing grid system cost is spread across fewer consumers.
The degradation of the grid started once the first intermittent generator was permitted to generate. Australia is almost uniquely placed for any household to generate and store intermittent power more economically than and gris supplier. Same solar panels are on grooves as those out in the fields. So no economy of scale yet the grid stuff is burdened with transmission and stability costs.
For those using capacity factor—it’s not useful in actual operations.
Both wind and solar have zero ( 0000!) availability every night and cloudy or rainy day for solar and days or weeks for wind.
So, make-up capacity (natural gas, oil, batteries) must be running or ready to run all the time—to make “renewables” almost equivalent to fossil fuel, nuclear, and hydro.
Of course, “renewables” should have their own make-up capacity on site so the rest of us don’t pay for it.
Then there are the other attributes of full time electricity that inverter generation cannot supply…should add that too.
From the post above by “michel” :
“Nothing” is a slight exaggeration, I prefer “A lot less than what people infer from the ‘installed capacity’ numbers provided by pro-renewable advocates”.
_ _ _ _ _ _
From the post by “nyeevknoit” :
Solar does indeed go to “zero every night”, but Wind is usually just in the “very low” range (less than 1 or 2 GW) for “days or weeks”.
_ _ _ _ _ _
While the month of May doesn’t show the extremes of “every winter” or “most summers”, the performance for the GB electricity grid so far this month illustrates many of the pitfalls of relying on “cheap and reliable … Honest gov’ner ! … renewables” for electricity generation.
Yes, that’s correct, and my phrasing was an exaggeration. But the problem is, you get some hours in those days or weeks when wind is down well below 1.0GW (from about 29GW faceplate). So you have to have duplicate in gas whatever you are counting on from wind, because it may not totally vanish, but its as close as makes no difference.
An hours shortfall will produce blackouts too, if not immediately remedied either by switching in the gas or shedding load.
What gets me is that they count hydro energy as “renewable” yet they’ve fought hydro (That is dams.) for decades. And they still want to remove dams.
Hydro has a track record of being reliable where it’s been feasible to build. (TVA, Niagara Falls)
It’s the new “renewables” that have a track record of not being reliable.
Hydro output should be removed from any calculations of the “success” of the new “renewables” energy output.
Well hydro is only semi reliable. Annual output in Norway varies between about 145TWh and 105TWh, depending on the precipitation and snowmelt as one example. Sometimes they’ve had to import largeish volumes from their neighbours, and others they have surplus to sell.