Guest Post by Willis Eschenbach [See update at the end]
We’ve been told over and over by very serious people that wind and solar are far cheaper than natural gas and coal … and yet the more renewables we add to the grid, the more our electricity bills keep going up and up. Ever wonder why that is?
If you want to see how to lie with numbers while wearing a suit and a straight face, look no further than Lazard’s Levelized Cost of Energy—LCOE, for those fond of acronyms and allergic to reality. LCOE is the financial world’s favorite energy cost yardstick: strip-mined of context, it tells you—supposedly—how much it costs to generate a megawatt-hour from any shiny new wind turbine, solar panel, or gas plant, averaged over its life.
Currently, Lazard says:
“Despite facing macro challenges and headwinds, utility-scale solar and onshore wind remain the most cost-effective forms of new-build energy generation on an unsubsidized basis (i.e., without tax subsidies). As such, renewable energy will continue to play a key role in the buildout of new power generation in the U.S. as the lowest-cost and quickest-to-deploy generation.”
See below how much cheaper wind and solar are than gas or coal? Sounds fair. Appears to be precise.
In reality? It’s financial alchemy, the spreadsheet version of Schrödinger’s Cat: deeply misleading, and possibly dead when you open the box.
Here’s the pitch. LCOE claims to offer a one-stop number: capital, operations, maintenance, fuel, and some sparkling optimism about the plant actually running at the output you wish it did.
What it doesn’t tell you—because the truth is a budget line item nobody wants to explain—is what it costs to actually plug that source into the chaos of a real grid. LCOE blithely ignores grid integration, backup, balancing, transmission upgrades, stabilization, wind turbine blades snapping off, and the nasty habit of solar and wind to give you exactly zero when you need power most.
In short, the LCOE measures the cost of what the plant should produce at the fence, not what it costs to deliver real, reliable electricity to your coffee maker when you actually want it.
Of course, to avoid lawsuits, in the fine print underneath the above graph Lazard says (emphasis mine):
Other factors would also have a potentially significant effect on the results contained herein but have not been examined in the scope of this current analysis. These additional factors, among others, may include: recent tariff-related cost impacts; implementation and interpretation of the full scope of the IRA; economic policy, transmission queue reform, network upgrades and other transmission matters, congestion, curtailment or other integration-related costs; permitting or other development costs, unless otherwise noted. This analysis is intended to represent a snapshot in time and utilizes a wide, but not exhaustive, sample set of Industry data. As such, we recognize and acknowledge the likelihood of results outside of our ranges. Therefore, this analysis is not a forecasting tool and should not be used as such given the complexities of our evolving Industry, grid and resource needs. Except as illustratively sensitized herein, this analysis does not consider the intermittent nature of selected renewables energy technologies or the related grid impacts of incremental renewable energy deployment. This analysis also does not address potential social and environmental externalities including, for example, the social costs and rate consequences for those who cannot afford distributed generation solutions.
and:
Variations in fuel prices can materially affect the LCOE of conventional generation technologies, but direct comparisons to “competing” renewable energy generation technologies must take into account issues such as dispatch characteristics (e.g., baseload and/or dispatchable intermediate capacity vs. peaking or intermittent technologies)
So, since Lazard is lying to us, what do we need to make informed decisions? Enter LFSCOE, the Levelized Full System Cost Of Energy—a metric so drearily comprehensive it’s almost honest. LFSCOE stuffs at least most of those hidden extras back into the variable column: the new transmission lines, the battery storage, the backup generators running on gas or coal, the cost of running a grid that isn’t just a science fair project for intermittent electrons. You want to compare wind, solar, nuclear, and gas on a somewhat level field? This is where the rubber finally meets the road, and it’s filled with potholes.
Here are some estimates. The LCOE for utility-scale solar? Lazard currently says $0.024–$0.096 per kilowatt-hour (kWh), which is $24 to $96 per megawatt-hour (MWh)—a figure recited like gospel at every green energy conference. For onshore wind, $0.024–$0.075/kWh, reported with the certainty of Newtonian physics. But plug these into the grid at scale and watch the magic unravel. When you add the hard system costs—grid balancing, expanded transmission, dedicated backup—real-world LFSCOE numbers emerge, and the story turns bleak for “cheap” renewables.
Recent system studies (IEA, EIA, Fraunhofer, you name it) peg the true LFSCOE for new onshore wind, at modest penetration levels (say, Germany or Texas): $0.08–$0.14/kilowatt-hour ($80 – $140 per megawatt-hour) after you factor in balancing, storage, congestion, and all the rest. Solar PV, barring the magical desert utopia where the sun shines every day, runs $0.07–$0.13/kWh if you want the lights to actually stay on and the grid not to wobble. Offshore wind, king of bad surprises, lopes in around $0.12-$0.18/kWh, and someone’s still writing up the bill for winterizing those turbine-servicing fleets of maintenance vessels.
Meanwhile, the real-world LFSCOE for new combined-cycle gas—dispatchable, flexible, can be sited anywhere, shows up during the Super Bowl—still hovers around $0.05–$0.075/kWh, system costs included. Nuclear? If you actually build it on budget (a rare unicorn), $0.09–$0.12/kWh, but the output’s reliable, the US is easing the regulatory mania which will lower prices, prices are dropping for the new mini-reactors, and you don’t have to build Texas-sized batteries for a cloudy week in February. Here’s an overview of the above numbers.
Lazard’s LCOE, in other words, is a metric for making renewables look good in PowerPoints, Excel dashboards, and glossy investor prospectuses. It’s the swimsuit edition of the energy-cost beauty contest—ignoring the fact that when you take solar and wind home to dinner, you’re also on the hook for their utility bills, therapy sessions, and spare bedrooms for all their unreliable friends.
It gets shadier. Lazard doesn’t wear the blame alone; the entire consultancy-policymaker-energy-industrial complex has played along. Banks issue green loans, politicians issue green press releases, and somewhere along the line, actual grid operators are left counting the hours until the next rolling blackout—brought to you by “record renewable penetration.”
So, does LCOE tell us anything useful?
Sure. It tells you what it would cost to build a power plant if you never had to connect it to anything, if electrons were wishful thinking, and if the world were as flat as a spreadsheet.
On the other hand, if you want to know what you’ll actually pay to keep the grid running—on a steamy Saturday afternoon in August in the Florida Panhandle, when it’s overcast and wind’s at a dead calm, everyone’s aircon sounds like it’s warming up for takeoff, and it’s anyone’s guess if your freezer will survive—you’d better reach for LFSCOE, dust off your reality glasses, and start counting all your hidden costs.
If energy policy were ever honestly debated, LCOE would be confined to the footnotes, right next to unicorn sightings and promised Powerball payouts.
But as long as reality is subordinate to narrative and the cost of electrons is calculated by the people selling you the next miracle, the first thing to get blacked out will be the truth.
Sigh …
My very best to all,
w.
Yeah, You’ve Heard It Before: When you comment, quote the exact words you are discussing. It’s surprising how much unnecessary friction it avoids.
[UPDATE] From a commenter in the thread below:
I didn’t say it was limited to a single technology, I said it models scenarios with a single technology being used for almost 100% of generation.
The part you are overlooking is that all of those additional costs come into play, not just at almost 100%, but as soon as you add unreliables to the grid.
For example. If you want to add a gigawatt of wind power to an existing grid, you absolutely have to add an additional gigawatt of conventional power for the dunkelflaute, as the Germans found out to their sorrow. Modern power grids only have backup spinning reserve for what they’re generating, so they can’t cover the new gigawatt.
And that nearly doubles the cost of that gigawatt, whether that gigawatt is 100% or 10% of your generation …
In addition, it starts on day one, not at 100%, and it’s not counted in the LCOE.
Next, you have the land costs because a wind farm covers a far greater area than conventional generation. And while the land won’t be expensive because it’s likely out in the boonies, it’s a real cost that starts on day 1 …
… plus, because it’s out in the boonies, the power is being generated where it’s NOT needed. That means that to get it where it IS needed you’ll have to put in roads, giant towers, high-voltage powerlines, transformers, and substations over a great distance.
And that’s expensive, and has to start well before day 1 … whether that gigawatt is 100% or 10% of your generation.
Etc. ad lib. Yes, as you imply, it must slightly overstate the costs at 10%, because the costs increase percentagewise the more unreliables you add to your grid …
… but that’s a bad thing, not a good thing …
Next, you say.
” It’s not superior to LCOE “
Including those very real costs and others beyond the scope of this comment is exactly why LFSCOE is far superior to LCOE. When I go to buy something, I want every single cost factored in.
The part you’re not acknowledging is the use of the LCOE by climate alarmists to convince people that renewable = cheap, when the reality is, almost every real-world example says renewable = expensive. It is misleading, which is bad enough. But that misleading claim is being deceptively used to aggressively sell a green agenda to an unsuspecting public, which is far worse. It’s bad data wrongly used in a worse cause.
Finally, the goal of the climate alarmists seems to be to have a wholly or almost wholly renewable grid … so LFSCOE would seem to be the proper choice in any case for this kind of long-term power supply question, no?
My best regards to you,
w.
I’ll tell you what: when unreliables turn sour most of the blame will be put on the grid operators by not providing enough capacity. Grid operators are used to dealing with projections of x demand and know how a reliable grid functions. But presented with highly dubious assumptions and assertions they are rather in limbo land. All they hear is MORE and FASTER. But they actually need to know the precise details, especially when faced w a two tier system. If i were a grid operator i would but strict limits to the amount of solar and wind into the grid.
Yes if I were a grid operator I’d limit it to zero.
Thee chart claims solar plus battery is less expensive than coal. How can someone look at that and not wonder why power companies want to build coal plants?
Many do exactly that. But those who wonder are not checking the math for themselves.
Amazing how big the blow-outs get in Australia at the moment. I expect the contractor/recipient is happy but users (us) end up with the bill. Another Labor government renewable project blows out by billions https://www.skynews.com.au/australia-news/politics/renewable-energy-zone-orana-blows-out-costs-to-55b-from-650m-adding-to-list-of-labors-net-zero-failures-across-regional-australia/news-story/a2f8614d1d421a478f8ba643ed961045
When calculating the value of the energy being produced, does LFSCOE take into account whether the energy being produced is actually needed? Or does it use a flat rate, the LCOE does.
Willis,
the statement that “Lazard is lying to us” appears to be as much a lie as what you are claiming
Lazard is doing. If you wanted to build a power station in the US and maximise your profits selling
electricity then you could use Lazard’s chart to decide what sort of power station to build. Since the transmission costs etc are completely seperate and borne by different companies why would you take them into consideration?
Futhermore the additional costs that you mention are “one-offs”. If the grid has been reconfigured to deal with renewable resources then any new renewable power sources would not incur them and so they should not be included in the generator costs. The costs are real and significant but they are not part of the cost of generating electricity and as such shouldn’t be included it figures.
Since the transmission costs etc are completely seperate and borne by different companies why would you take them into consideration?
The cost to society and the environment from ruinables (unlike coal) is astronomical, you twonk. The evidence is everywhere. What the hell is wrong with you people?
You guys run some weird arguments
Point 1.) So now if you just break the costs apart to a separate company then you don’t count it. Okay so on my coal fired power station it buys coal from a different company so we don’t count the cost of buying the coal. So my generation cost is now just the cost to run the plant and boy is that going to be a low figure because my staff are in another company contracted to run it so I am saying my cost is close to $0 per kWh.
Point 2.) The grid changes are not a one off there is an ongoing cost for things like grid batteries. Those things only exist on the grid for the renewables and now you want to treat them as just normal.
You keep ignoring the fact you are arguing to replace a perfectly good working system. So lets go back to the good old days and we have horse power and gas lights. Your argument would go the cost of the new fangled electric system and fossil fuel car and road system shouldn’t count making tar roads and installing poles and cables as a cost because it’s just like the dirt roads and gas pipes of the current systems in place 🙂
Leon,
the costs are real. The question is who has to pay for them. If a country has split off power generators companies from transmission companies then clearly the power generators have zero interest or concern on the effects that their generators have on the grid and will act to maximise their profits. Alternatively in a country where the entire electricity system is state owned then there will be a different set of concerns and they would want to minimise the total cost. Which metric you use depends on the how the electrical grid works in a particular region.
The issue is not the metric it is the fact that people use the wrong metric when measuring different things.
“Alternatively in a country where the entire electricity system is state owned then there will be a different set of concerns and they would want to minimise the total cost.”
What world do you live in?
State owned is command economics and tax payer money. Supply and demand, free market economics is where people want to reduce or minimize costs.
It’s not hard to work out which countries follow LFSCOE and those who follow LCOE. It is also not hard to work out which countries have the higher GDP as a result of the former policy. Australia’s declining per capita wealth is directly attributable to the costings associated with LCOE while ignoring the real role of LFSCOE across the economy. Added to that is the rank stupidity of the 3-mine uranium policy and the banning of nuclear energy.
So far capital has fled the hydrogen market, dimmed its lights on the offshore wind market. May the trend continue to onshore wind factories and to solar factories.
Well, the real problem with this stuff is that the activists want to invent some new way of doing capital investment appraisal just for renewables. Its idiotic, these are just investment proposals like any others, and there is an accepted methodology for doing the appraisal, its called Net Present Value analysis, and you can find it explained at length in any Corporate Finance textbook. I recommend Brealey and Myers, but there are shelf loads of them.
None of this other stuff is worth a damn.
To compare two systems with different technologies all you do is write down by year the total cash flows out required to deliver the same product to the same market. The total flows! The same product….! Then apply to these cash flows the NPV function from the spreadsheet of your choice. The one with the lower NPV is the least expensive.
The difficulty with doing this, and the reason its a large subject, is not the method or the formula. Its the plethora of fallacies people fall into while calculating what are the cash flows to discount, and the fallacies they fall into while trying to include things which are not cash flows at all. So you will find lots of cautionary tales in the textbooks and explanations of how to handle different scenarios to get to proper cash flows.
But in the end, the way to think about LCOE and its derivatives is that these are complicated ways of trying to avoid doing standard business case financial appraisals, and the answer to them is just to stop it, and do it right.
Its like Nick wants to do cost accounting by looking only at the fuel costs of two generating systems, and thinks that the one with lower fuel costs must be the low cost producer. Like that water powered mills must be cheaper than steam powered ones, since the stream water is free, it just falls from the sky. Like that sailing ships must be cheaper freight transport than steam, because the wind is free. Look Ma, no fuel!
When people find it necessary to invent new accounting systems to justify their pet projects, you know there is something wrong, and its not traditional accounting, its the projects.
A bit like inventing new statistical methods to draw hockey sticks…
In Australia, our ‘energy’ minister, Blackout Bowen, was asked in parliamentary question time for the cheapest energy source to the consumer.
Several times he responded by quoting generator costs, as is typical and noted by Willis.
Lying by omission is still lying and is supposed to be illegal in our parliamentary system but never prosecuted, unfortunately
My electricity supplier is always asking if I wish to utilise renewable, green, energy sources
For supposedly cheaper energy, they also require an extra cost/kWh for this Earth-saving technology
Thanks, John. You inspired me to do a little research.
In the US, about 850 electrical utilities offer a “green” or “low-carbon” option to the consumer.
Almost all of them charge for the privilege.
The typical charge is about one to two cents per kWh, with the high end being about seven cents per kWh.
Presumably, this is because renewable energy is so d@mn cheap …
w.
Based on what I have seen and experienced, it is a middleman cost/fee.
Guaranteed renewable electricity is always higher priced. It’s offered as a way to feel virtuous about paying your electrical bill. The PUC allows for this only when actual cost schedules are submitted to them in advance for approval.
So it definitely looks that there are two methods for determining renewable electricity prices. One way is the way liars try to sell stuff and the other way is what the PUC actually permits the rates to be.
Very nice Willis. What gravels me is that fossil fuels bear the cost of staying on stand by and filling in for renewables when they fail. All costs to stay on stand by and fill in the gap when renewables fail should be paid for by renewables and included in their cost of operation. Renewables can not sustain the grid, they must be removed from the grid and all fossil fuel and nuclear fired and run at their cost effective levels. Build more fossil fuel and nuclear.
Don’t worry. When the EPA rescinds the 2009 CO2 Endangerment Finding and Trump finally terminates subsidies for the renewables, these will fall by wayside.
Math is hard.
“Meanwhile, the real-world LFSCOE for new combined-cycle gas—dispatchable, flexible, can be sited anywhere, shows up during the Super Bowl—still hovers around $0.05–$0.075/kWh, system costs included. Nuclear? If you actually build it on budget (a rare unicorn), $0.9–$0.12/kWh, but the output’s reliable, the US is easing the regulatory mania which will lower prices, prices are dropping for the new mini-reactors, and you don’t have to build Texas-sized batteries for a cloudy week in February. “
Sorry to be picky, but surely you meant $0.09-$0.12 kWh.
Thanks for being picky, Andrew. I hate typos. Best part about writing for the web? Folks like you who don’t let my errors last long.
w.
Rather than look at COSTS for intermittent power like PV and wind, consider the VALUE of adding non-dispatchable power to a fully functional power grid. The need for reliable standby power means their capital, maintenance and operating costs remain the same. At most, intermittent power only reduces fuel consumption. In practice, the additional costs Willis described reduce the value even further.
Nicely done, Willis.
Perhaps you can arrange to havie it tattooed backwards on Nick’s forehead so he can read it every time he looks in the mirror.
Seems like a total waste of taxpayer subsidy if it promotes something that causes prices to go up and eliminates the competition. On the other hand, that is definitely progressive government.
Dear Willis,
As far as I can tell, the references you cite discuss LCOE, but not LFSCOE. I reviewed the most recent Fraunhofer studies and the EIA 2023 levelized cost study you cite. All sing the praises of low-cost wind and solar.
Perhaps the links are incorrect? Or have I misread your article?
Thanks and keep up the wonderful work!
Bottom line-what underpins our entire civilization is cheap, reliable energy. Instead of producing food and products with energy produced by our bodies, many times more energy is obtained from other sources. This is why the planet supports around 8 billion people rather than around 8 million in the Neolithic era, just 10,000 years ago. By going backwards in terms of energy production to then put in to food and goods, populations MUST collapse.
All this potential catastrophic collapse of our species due to the coercive efforts of a few vested interests who evade any discussion and production of hard evidence to prove their narrative. Humans are actually pretty dumb.
Suckers for emotional appeals in many cases.