Solar and wind are low density power systems and work 20-35% of the time. The rest of the time you need backup. You can't replace a coal/gas/nuclear plant (say 1 Gwatt plant) with 1 Gw of solar (or wind) you need to build 3-4 Gw, and then at least 12 hours of battery or some sort of backup. A review of California, UK or Canadian minute by minute power inflow and demands show that every now and then, 2-6x/year there will be days when no wind blows and essentially no sun shines. If you plan for these you will need more than 12 hours of backup. So even at $1/watt for utility solar build costs your built out system capex, ignoring the transmission lines will result in a much higher capex for the system than exists for the present coal/gas/nuclear/hydro system, which means the $ cost per kw-hr is not going to be anywhere near "so cheap it will cause an economic boom" scenario. Ask Californians what price fluctuations they see, and ask UK and German rate payers what there kw-hr costs are now versus 20 yrs ago. And you ignore the nimby factor and the environmental impact and regulatory regime that will greatly hinder any major solar/wind/transmission line build out unless we have a political revolution.
Naive question..how "competitive" would solar be if the panels weren't being build by Uyghur slave labor and other CCP subsidies?
I'm less enthusiastic about solar also because even if you ignore the above issue, there are unanswered questions regarding massive disposal of degraded solar panels, the enormous amount of land usage needs of industry and the environmental impact on 3rd world countries from the mining for the rare earth minerals needed for a solar economy.
Can we just do Gen III or Gen IV nuclear power? We know that fission can deliver on demand power to industrial society. Solar is still a question mark to me
The energy thesis is not new, see "Where is My Flying Car?" by J. Storrs Hall.
That is not a refutation of the "regulation caused the Great Stagnation" thesis, it's actually a vindication. Regulation choked nuclear and is the reason it's so expensive.
That's also a key reason why Jim Pethokoukis & others like J. Storrs Hall are less excited about solar, they are afraid regulation did not create a level playing field.
I'm personally agnostic, I'd like to have more of everything and a level-playing field for all technologies - but I do share concerns like ...
- if solar is so great, why aren't we getting it faster? From a cost perspective, it disappointed strongly in the late 2000s; I hear my friends in renewable energy talking about it for a decade and little happens.
- the LCOE analyses has been criticised for not properly comparing, e.g. nuclear is artificially expensive and solar & wind artificially cheap (I can't adjudicate whether that's true or not)
- Solar & wind suffer often from the same problems that other sources of energy like nuclear suffer (albeit less of those): NIMBYism, land use regulation, environmental reviews etc.
So we definitely need to do the work on de-regulation or "energy innovation zones" (special economic zones with less regulation), a nuclear revival and a new science race to find other cheaper sources, even better battery technology etc. either way.
Betting just on existing solar technology doesn't seem enough.
As usual, the original post is focused solely on "green" answers solely from an American perspective. In and of itself, that is not unexpected or unrealistic. However, the true costs on a global scale (especially to the countries that produce the basic materials for batteries) are never factored in. True, American energy costs are falling, and the applied formulas seem to make sense on the surface. But, I would be more accepting if a worldwide perspective were taken. The "bloom" is rapidly coming off the "green" rose around the world as true costs are being realized. I realize this is unsubstantiated for the moment, but if a conversation ensues I will gladly provide required
'Jim Pethokoukis, for example, who writes my favorite techno-optimist blog, pointedly refuses to mention solar at all.'
Because solar is coded left, an example of where the dirty hippies were right and the 'realist' right aligned types were wrong. Much better to hype up nuclear, where he can blame the hippies and government regulation for killing it. Ignore the fact that it's currently failing the market test.
I like Jim Petholoukis' blog too. But a nuclear plant requires 500-800 staff fulltime (NEI). That's a lot of opex. It is hard to see how nuclear will ever be really cheap unless you address both the opex and the capex question (I suspect the cheapness of nuclear in the 1950s was due to a combination of subsidies, loss leader pricing and lax safety requirements).
Meanwhile here in Australia now one third of all households now have some solar on the roof. Just deregulate solar on the roof. Two guys on the roof for a day then you're set for twenty years.
IMHO people over-dramatise the challenges of intermittent solar. David Osmond posts regular model updates showing Australia can get by with close to 100% renewables and only 5 hours storage (https://twitter.com/DavidOsmond8). Many people think Australia will be able to get more than that from the light vehicle fleet alone.
This is oversimplified (haven't even discussed wind, hydro and hydrogen). But by 2030 a lot of people will live in hot countries with a lot of EVs on the street. And cheap solar + bidirectional chargers can get you surprisingly far with no megaprojects required.
Isn’t part of the problem the amount of non-renewables it takes to build wind and solar? Maybe I’m out of the loop, but doesn’t it offset at the very least as things are? I am trying to get on the renewables train, but it feels like the main solution is to address poverty in those places where pollution is a consequence of economic hardship. Maybe it’s a both and I’m missing things? Anyone care to educate me?
In other news, I wrote an article about Andrew Tate and the internet fathers of today. I’m an aspiring writer trying to hone my craft. Any eyeballs and input matters!
Without going all the way down the nuclear rabbit-hole I'm still increasingly convinced that solar/wind aren't viable as a whole-grid.
The problem loosely is that because they are intermittent (and as a civilisation we can't really tolerate not having power) they need to be both significantly overbuilt and have something approximating a 100% back-up. Whether that's a whole fleet of gas plants or a monumental amounts of batteries. The cost of the over-build and the way the cost of the back-up needs to be added to the solar cost. When you do that it becomes obvious that it isn't really viable as a whole-grid.
I'm disappointed that the author didn't address the resource issue related to solar and wind power: is there enough lithium, are there enough rare-earth metals on the planet to support the battery requirements of these cheap but partial energy sources, on the scale needed to power this abundant energy future, for 7 billion people and counting? Do we think we're going to get batteries out of rocks? Battery technology is constrained by physics; available land is constrained by geography. These are real issues, I think, that should be addressed in a techno-optimist piece?
Energy expenditures are <5% of US GDP. Any productivity is a good thing but this feels like a drop in the bucket compared to the large (and growing) sources of anti-productivity: housing, healthcare, education, etc.
I share your optimism on energy but at the same time fear it gives us a false sense that we are beginning to solve the productivity problem in aggregate. We are not (wake me up when we fix zoning).
" cheap, energy-dense batteries". Until these batteries become commercially available, solar and wind will remain significantly more expensive than combined cycle gas (especially offshore wind).
* 365 days per year = 438 kWh per year per square meter
Round numbers: a 1 meter square set of panels will generate about 400 kWh of electricity per year.
Total US usage: 3.8 trillion kWh
Round numbers, call it 4T.
4T kWh / 400 kWh = 10B m^2
This is the amount of land area that would have to be covered with solar panels in order to provide for our current electricity usage. Being America, I have a hard time relating to square meters, so how big is 10 billion m^2? An acre is about 4,000 square meters. So...
10B m^2 = 2.5M acres (+/-) = 4000 sq miles (+/-)
How big is that? It's about the size of the Baltimore / Washington DC metroplex. Paving over this area with solar panels would therefore improve life both by what is destroyed and by what it produced. We may need to enlarge it to accommodate the coming electrification of the entire transportation system. (Of course, with DC's bureaucratic apparatus paved in silicon and lithium, that may not happen, but I digress...) I think covering most of New Jersey would take care of that too. Leave the Jersey Shore; its reality TV is far too entertaining.
Oh... and expect to replace all of this every 20-25 years.
---
To be clear, while I disagree with your solar optimism, Noah, I love this article, since it points out the critical importance of energy cost to everything we do. It is not hyperbole to say that human developmental history IS the gradual discovery and cultivation of new and more dense energy sources. Any eco-movement that fails to recognize this is shooting itself in the foot.
Personally, my money is on the fusion people. You know what they say: "fusion energy is just 10 years away!" Unfortunately, "they"' have been saying this for my entire 50 year life, but I still have hope. Thanks for another great article.
Yeah.... batteries with elements dug up by slaves? Solar panels made by slaves? And to know they take up so much land...and not to forget that they’re not recyclable? Cheaper to make new ones? Also lose their strength year over year? Nah.
Noah, when you cite Lazard LCOE as proof that solar is now the cheapest form of electricity, you’re simply telling us all that you haven’t done a realistic, in-depth analysis.
From Lazard’s website: “Analysis excludes integration (e.g., grid and conventional generation investment to overcome system intermittency) costs for intermittent technologies.”
Also from Lazard: “Although alternative energy is increasingly cost-competitive and storage technology holds great promise, alternative energy systems alone will not be capable of meeting the base-load generation needs of a developed economy for the foreseeable future. “
When you flip on your light switch, the product you access is reliable on-demand 24x7 electricity. This is not the product exiting a solar array field.
While it's exciting to think of a battery centric future there isn't enough Lithium in the world to run it for cars alone, not to mention backing up excess solar production. When we can get solid state batteries using different, cheaper metals I think we'll 'be there'. In the meantime we should be building SMRs and getting thorium plants figured out (we can burn existing nuclear waste stockpiles in thorium plants!). I'm not trying to malign solar--its great for some things (desal is perfect if they can figure it out better as it produces when the sun's out and doesn't matter if it shuts down at night). But we need constant power sources and the current SMR designed nukes are the best replacement for large scale fossil fuel generation--modular, simple, and safe.
Solar and wind are low density power systems and work 20-35% of the time. The rest of the time you need backup. You can't replace a coal/gas/nuclear plant (say 1 Gwatt plant) with 1 Gw of solar (or wind) you need to build 3-4 Gw, and then at least 12 hours of battery or some sort of backup. A review of California, UK or Canadian minute by minute power inflow and demands show that every now and then, 2-6x/year there will be days when no wind blows and essentially no sun shines. If you plan for these you will need more than 12 hours of backup. So even at $1/watt for utility solar build costs your built out system capex, ignoring the transmission lines will result in a much higher capex for the system than exists for the present coal/gas/nuclear/hydro system, which means the $ cost per kw-hr is not going to be anywhere near "so cheap it will cause an economic boom" scenario. Ask Californians what price fluctuations they see, and ask UK and German rate payers what there kw-hr costs are now versus 20 yrs ago. And you ignore the nimby factor and the environmental impact and regulatory regime that will greatly hinder any major solar/wind/transmission line build out unless we have a political revolution.
Naive question..how "competitive" would solar be if the panels weren't being build by Uyghur slave labor and other CCP subsidies?
I'm less enthusiastic about solar also because even if you ignore the above issue, there are unanswered questions regarding massive disposal of degraded solar panels, the enormous amount of land usage needs of industry and the environmental impact on 3rd world countries from the mining for the rare earth minerals needed for a solar economy.
Can we just do Gen III or Gen IV nuclear power? We know that fission can deliver on demand power to industrial society. Solar is still a question mark to me
The energy thesis is not new, see "Where is My Flying Car?" by J. Storrs Hall.
That is not a refutation of the "regulation caused the Great Stagnation" thesis, it's actually a vindication. Regulation choked nuclear and is the reason it's so expensive.
That's also a key reason why Jim Pethokoukis & others like J. Storrs Hall are less excited about solar, they are afraid regulation did not create a level playing field.
I'm personally agnostic, I'd like to have more of everything and a level-playing field for all technologies - but I do share concerns like ...
- if solar is so great, why aren't we getting it faster? From a cost perspective, it disappointed strongly in the late 2000s; I hear my friends in renewable energy talking about it for a decade and little happens.
- the LCOE analyses has been criticised for not properly comparing, e.g. nuclear is artificially expensive and solar & wind artificially cheap (I can't adjudicate whether that's true or not)
- Solar & wind suffer often from the same problems that other sources of energy like nuclear suffer (albeit less of those): NIMBYism, land use regulation, environmental reviews etc.
So we definitely need to do the work on de-regulation or "energy innovation zones" (special economic zones with less regulation), a nuclear revival and a new science race to find other cheaper sources, even better battery technology etc. either way.
Betting just on existing solar technology doesn't seem enough.
As usual, the original post is focused solely on "green" answers solely from an American perspective. In and of itself, that is not unexpected or unrealistic. However, the true costs on a global scale (especially to the countries that produce the basic materials for batteries) are never factored in. True, American energy costs are falling, and the applied formulas seem to make sense on the surface. But, I would be more accepting if a worldwide perspective were taken. The "bloom" is rapidly coming off the "green" rose around the world as true costs are being realized. I realize this is unsubstantiated for the moment, but if a conversation ensues I will gladly provide required
'Jim Pethokoukis, for example, who writes my favorite techno-optimist blog, pointedly refuses to mention solar at all.'
Because solar is coded left, an example of where the dirty hippies were right and the 'realist' right aligned types were wrong. Much better to hype up nuclear, where he can blame the hippies and government regulation for killing it. Ignore the fact that it's currently failing the market test.
I like Jim Petholoukis' blog too. But a nuclear plant requires 500-800 staff fulltime (NEI). That's a lot of opex. It is hard to see how nuclear will ever be really cheap unless you address both the opex and the capex question (I suspect the cheapness of nuclear in the 1950s was due to a combination of subsidies, loss leader pricing and lax safety requirements).
Meanwhile here in Australia now one third of all households now have some solar on the roof. Just deregulate solar on the roof. Two guys on the roof for a day then you're set for twenty years.
IMHO people over-dramatise the challenges of intermittent solar. David Osmond posts regular model updates showing Australia can get by with close to 100% renewables and only 5 hours storage (https://twitter.com/DavidOsmond8). Many people think Australia will be able to get more than that from the light vehicle fleet alone.
This is oversimplified (haven't even discussed wind, hydro and hydrogen). But by 2030 a lot of people will live in hot countries with a lot of EVs on the street. And cheap solar + bidirectional chargers can get you surprisingly far with no megaprojects required.
Isn’t part of the problem the amount of non-renewables it takes to build wind and solar? Maybe I’m out of the loop, but doesn’t it offset at the very least as things are? I am trying to get on the renewables train, but it feels like the main solution is to address poverty in those places where pollution is a consequence of economic hardship. Maybe it’s a both and I’m missing things? Anyone care to educate me?
In other news, I wrote an article about Andrew Tate and the internet fathers of today. I’m an aspiring writer trying to hone my craft. Any eyeballs and input matters!
https://open.substack.com/pub/tothestorehouse/p/abdicated-responsibility?r=tn513&utm_medium=ios&utm_campaign=post
Without going all the way down the nuclear rabbit-hole I'm still increasingly convinced that solar/wind aren't viable as a whole-grid.
The problem loosely is that because they are intermittent (and as a civilisation we can't really tolerate not having power) they need to be both significantly overbuilt and have something approximating a 100% back-up. Whether that's a whole fleet of gas plants or a monumental amounts of batteries. The cost of the over-build and the way the cost of the back-up needs to be added to the solar cost. When you do that it becomes obvious that it isn't really viable as a whole-grid.
This is some interesting analysis (albeit potentially too pessimistic, I haven't unpacked the model) https://jackdevanney.substack.com/p/nuclear-and-windsolar
I'm disappointed that the author didn't address the resource issue related to solar and wind power: is there enough lithium, are there enough rare-earth metals on the planet to support the battery requirements of these cheap but partial energy sources, on the scale needed to power this abundant energy future, for 7 billion people and counting? Do we think we're going to get batteries out of rocks? Battery technology is constrained by physics; available land is constrained by geography. These are real issues, I think, that should be addressed in a techno-optimist piece?
Energy expenditures are <5% of US GDP. Any productivity is a good thing but this feels like a drop in the bucket compared to the large (and growing) sources of anti-productivity: housing, healthcare, education, etc.
I share your optimism on energy but at the same time fear it gives us a false sense that we are beginning to solve the productivity problem in aggregate. We are not (wake me up when we fix zoning).
" cheap, energy-dense batteries". Until these batteries become commercially available, solar and wind will remain significantly more expensive than combined cycle gas (especially offshore wind).
Solar panel production:
150 W/sq m * 8 hours sun = 1.2kWh / m^2 / day
* 365 days per year = 438 kWh per year per square meter
Round numbers: a 1 meter square set of panels will generate about 400 kWh of electricity per year.
Total US usage: 3.8 trillion kWh
Round numbers, call it 4T.
4T kWh / 400 kWh = 10B m^2
This is the amount of land area that would have to be covered with solar panels in order to provide for our current electricity usage. Being America, I have a hard time relating to square meters, so how big is 10 billion m^2? An acre is about 4,000 square meters. So...
10B m^2 = 2.5M acres (+/-) = 4000 sq miles (+/-)
How big is that? It's about the size of the Baltimore / Washington DC metroplex. Paving over this area with solar panels would therefore improve life both by what is destroyed and by what it produced. We may need to enlarge it to accommodate the coming electrification of the entire transportation system. (Of course, with DC's bureaucratic apparatus paved in silicon and lithium, that may not happen, but I digress...) I think covering most of New Jersey would take care of that too. Leave the Jersey Shore; its reality TV is far too entertaining.
Oh... and expect to replace all of this every 20-25 years.
---
To be clear, while I disagree with your solar optimism, Noah, I love this article, since it points out the critical importance of energy cost to everything we do. It is not hyperbole to say that human developmental history IS the gradual discovery and cultivation of new and more dense energy sources. Any eco-movement that fails to recognize this is shooting itself in the foot.
Personally, my money is on the fusion people. You know what they say: "fusion energy is just 10 years away!" Unfortunately, "they"' have been saying this for my entire 50 year life, but I still have hope. Thanks for another great article.
Solar and wind should be called unreliables. Sorry...
Yeah.... batteries with elements dug up by slaves? Solar panels made by slaves? And to know they take up so much land...and not to forget that they’re not recyclable? Cheaper to make new ones? Also lose their strength year over year? Nah.
Noah, when you cite Lazard LCOE as proof that solar is now the cheapest form of electricity, you’re simply telling us all that you haven’t done a realistic, in-depth analysis.
From Lazard’s website: “Analysis excludes integration (e.g., grid and conventional generation investment to overcome system intermittency) costs for intermittent technologies.”
Also from Lazard: “Although alternative energy is increasingly cost-competitive and storage technology holds great promise, alternative energy systems alone will not be capable of meeting the base-load generation needs of a developed economy for the foreseeable future. “
When you flip on your light switch, the product you access is reliable on-demand 24x7 electricity. This is not the product exiting a solar array field.
While it's exciting to think of a battery centric future there isn't enough Lithium in the world to run it for cars alone, not to mention backing up excess solar production. When we can get solid state batteries using different, cheaper metals I think we'll 'be there'. In the meantime we should be building SMRs and getting thorium plants figured out (we can burn existing nuclear waste stockpiles in thorium plants!). I'm not trying to malign solar--its great for some things (desal is perfect if they can figure it out better as it produces when the sun's out and doesn't matter if it shuts down at night). But we need constant power sources and the current SMR designed nukes are the best replacement for large scale fossil fuel generation--modular, simple, and safe.