Hybrid Battery Storage System (Image by Ysc Use)
Astronomical battery cost looms over “renewables”
By David Wojick
The amount of storage needed to make renewables reliable is so huge that even if the cost dropped fantastically, we still could not afford it.
We now know that the battery storage for the entire American grid is impossibly expensive, thanks to a breakthru study by engineer Ken Gregory. Looking at several recent years he analyzed, on an hour-by-hour basis, the electricity produced with fossil fuels. He then calculated what it would have taken in the way of storage to produce the same energy using wind and solar power. He did this by scaling up those year’s actual wind and solar production.
Based on his work, which only covered 48 states, our round working estimate of the required storage for the whole country is an amazing 250 million MWh. America today has less than 20 thousand MWh of grid scale battery storage, which is next to nothing.
Grid scale batteries today cost around $700,000 a MWh. For 250 million MWh we get an astronomical total cost of $175 trillion dollars just to replace today’s fossil fuel generated electricity needs with wind and solar. Even the fantastically low cost estimates that some people are proposing puts the cost around the total annual GDP of America. Even worse, if we get the electric cars and trucks the Biden Administration is calling for these astronomical numbers could easily double.
Note that the range of estimates for future battery unit cost is itself enormous, so no single figure is credible. Turns out that does not matter, because even the fantastically low estimates give fantastically high total costs.
Let’s start with the reality. The EIA has collected annual utility data on the cost of grid scale battery arrays. A recent report is “Battery Storage in the United States: An Update on Market Trends — August 2021”.
From 2013 to 2018 the average reported cost was around $1,500,000 per MWh. The range was pretty large, from under $500,000 to around $3,000,000 per MWh.
It is worth noting that in 2020 EIA excitedly reported a big drop in cost. This was from an average of $2,100,000 in 2015 way down to a low of $600,000 in 2019. I am rather skeptical that this 70% cost drop was real. There was no technological breakthrough to cause it. I suspect it was either a case of price cutting or of utilities manipulating their cost reports. Tesla has been bidding very low, at around $500,000 for some time. These are likely loss leader bids. Today Tesla charges about $650,000 per MWh just for the batteries, much less whole facilities.
It is certainly the case that the cost must be going way up these days, not down, given the huge price spike in lithium and other essential constituent materials, as well as in the energy needed for making these monster battery arrays.
So it seems fair to say that the cost is at least $600,000 a MWh, quite possibly a lot more. A million dollars a MWh is not an unreasonable estimate. Keep in mind that a MWh is what an average American home uses in just a month, so it is not a lot of juice storage for a lot of money.
Now comes the fantasy. There are several recent mainstream estimates of the future capital cost of grid scale battery arrays. These estimates are often used in fantasy assessments of the economic feasibility of a transition from coal and gas fired generation to wind and solar. The battery cost estimates are crucial because it will take an enormous amount of batteries to try to make intermittent wind and solar reliable.
For example, DOE’s National Renewable Energy Laboratory has published battery cost projections through 2050 in their report “Cost Projections for Utility Scale Battery Storage: 2021 Update”. NREL is gung-ho on renewables, so also on the batteries needed to try to make wind and solar reliable.
Each NREL projection is for a narrow range of costs. The low end of that range is a mere $143,000 per MWh in 2030 and $87,000 in 2050. That is right, just $87,000 for something that today costs $600,000 to $1,000,000, with costs going up.
Clearly this projection is extremely rosy, to the point of fantasy.
In a recent report — “The Future of Energy Storage” — MIT goes even lower. Their 2050 battery cost estimate is a tiny $70,000 per MWh! For something that costs upwards of half a million dollars today. Surely this is pure fantasy.
Mind you given the Biden goal of zero electric power emissions by 2035, the 2050 fantasy figure may be irrelevant since what counts is the cost between now and 2035. But even NREL’s 2030 estimate of $143,000 is unbelievable. Given the way prices are rising, $1,000,000 is a better bet. Biden’s goal is pure fantasy.
Energy policy needs to be based on sound engineering estimates, not wishful fantasies. But in this case even the fantastically low unit costs yield impossibly high total costs. This shows how truly impossible the renewables plus storage policy really is.
Listed below are the total storage costs for our 250,000,000 MWh, for each of the possible unit costs discussed above.
$1,500,000 per MWh = $375 trillion
$1,000,000 = $250 trillion
$600,000 = $150 trillion
Gregory’s $347,000 = $87 trillion
NREL $143,000 = $36 trillion
NREL $87,000 = $22 trillion
MIT $70,000 = $18 trillion
US annual GDP is about $23 trillion, so even the fantastically low MIT number gives a storage cost nearly equal to that enormous number. More reasonable unit costs quickly get the total into the hundred trillion dollar range or more.
Clearly these astronomical cost figures show that battery storage is simply impossible when it comes to making wind and solar reliable at grid scale. This is why the battery systems being built today are as nothing compared to what is actually needed to go with the wind and solar systems they accompany. Yet it is claimed that wind and solar plus batteries are replacing reliable fossil fuel power plants. This is simple and hugely false.
If wind and solar developers were required to include the batteries needed to make their projects reliable, none would ever be built. I invite that requirement.
David Wojick, Ph.D. is an independent analyst working at the intersection of science, technology and policy. For origins see http://www.stemed.info/engineer_tackles_confusion.html For over 100 prior articles for CFACT see http://www.cfact.org/author/david-wojick-ph-d/ Available for confidential research and consulting.