I don't think we're anywhere near maxing out the cost curve for solar. I think we may be close to maxing out the cost curve for traditional utility-scale installations, though. The cost of the panels themselves is a pretty small--maybe 20% of the final cost. The other cost is for things like the inverters, mounting hardware, labor, etc. But you can drive down the cost of thoses aspects quite a bit.
For example, you don't need an inverter. Lot's of applications, such as data centers, electrolyzers, or even HVDC transmission can operate on DC power. Another significant cost is mounting hardware, to keep the solar panels in an optimal position relative to the sun, either tracking or fixed. But you can also just drop the solar panels on the ground! You lose a bit of efficiency, but you can make up for that by using more solar panels. Oh, and this also cuts down by a huge amount on the labor needed to install the mounting hardware and mount the solar panels.
Some calculations on ChatGPT tells me that a single standard-sized container could hold about 10 MW of thin-film solar panels. You could conceivably manufacture a container-sized device that automatically unrolls 10 MW of solar power onto unused desert land with virtually zero mounting hardware, inverter, and labor costs, taking full advantage of economies of scale. There is a lot of space left on the learning curve to reach that point.
Isn’t the reason for AC that it makes voltage transformation easy? If you trade your inverters for more complex transformers … you might end up with a win, but it’s not clear to me you do.
Alternatives run everything with the same voltage (not great) or have multiple circuits with different voltages (probably also bad).
I can't for the life of me understand the Sun Cable proposal. I've worked in energy, but I'm not an engineer. Perhaps some smart folks can tell me I'm wrong, but besides some kind of "blue sky thinking" grift, how on earth does it make sense to link Darwin to Singapore with a power cable? The distance from DRW - SIN is greater than the distance from DRW - SYD. It's much greater than DRW - BNE, which would also be a more logical point to link Darwin to Australia's east coast electricity grid. Surely it's cheaper and less risky to build overland across Australia than laying undersea cable through (largely) Indonesian waters? Can it really work out cheaper laying deep sea cable to South East Asia rather than connecting Top End solar to other Australian locations? Is the power price differential, or demand level in Singapore, really large enough to justify such a cost and risk?
The only sensible part is the time difference - when regional Northern Australia is generating power at midday and afternoon, Singapore is coming online in the morning. Power going to East Coast Australia doesn't have this advantage I suppose. But still, midday power from Northern Australia would be selling into the evening peak on the East Coast.
It doesn't seem to make sense. It seems there is some bias in people's thinking where they assume that Singapore is close to Darwin, because they're all tropical, right? Now Australia's big, but so is the distance between Aus and South-East Asia.
Undersea data cables are actually considerably less expensive per mile than overland. I don't know if the same dynamic holds for power cables, but if so that would explain the apparent discrepancy in the plan.
To follow on my earlier point: many fibre optic cables connecting countries in South America and Africa actually just bunny hop across the coast, because that's where the population is and it's easier to build!
It's because Singapore wants to buy way more power than the rest of Australia does; this is very much a one-way cable in practice. I also assume (but might be wrong) that power sold in Singapore can make its way into Malaysia too.
The big advantage of the Northern territory is a big sunny desert, but other Australian states also have big sunny deserts of their own. SEAsia has rainfall and thus cloud cover, and way less empty open flat land.
Re: banning plastic bags; the various UK administrations have mandated a (very small- I think it's 5p) charge on disposable plastic shopping bags. I've no idea what the effect is on plastic use, but it certainly has improved the littering situation; places just look nicer with fewer plastic bags around the place. I mean, yes, a ban is worse than a tax, sure. But I'm not sure that I oppose that move.
I do oppose the ban on plastic straws (we have that in the UK, too); that is pretty unarguably silly. But only technically; I don't really care about it personally.
The tiny fees are in theory fine, but my experience is that transaction costs are substantial. If it actively did reduce the number of bags, the question is whether it was due to the imposed transaction costs or the actual fee.
I'm not convinved such bag fees actually decrease usage, or even littering...though of course the locales aren't quite comparable. My store doesn't do plastic bags anymore (one-off experimental thing during some Christmases and covid, some customers love them, most hate them and they're super annoying for us cashiers to fill), only paper, and we're supposed to charge...25 cents(!) per bag. That is, 50 cents for a double bag, which is the most common usage cause everyone's paranoid about singles breaking. People on foodstamps are exempt by law - talk about cross-pressure incentives - but in practice...we mostly just don't bother assessing the fee* at all. It feels super petty, customers hate it, the revenue goes 100% to the government, there's no accountability for collection, it slows down checkout flow, and most importantly it doesn't actually deter people from treating the bags as free and deserved. The most price-sensitive people will still carry all their stuff bagless, or improvise (free) produce bags as shopping bags, so it helps slightly on the margin. But these thrifters are not responsible for the bulk of bag consumption anyway! That's the families of 4 who spend hundreds of dollars and end up with dozens of bags per cart, as well as the buy-a-single-item kids who somehow need a bag for their bag of chips. Both are insensitive to the fee, the former because it's a rounding error, and the latter because they're almost always paying with Someone Else's Money so who cares. If such a mandate can't work in SF, I don't expect it to make a difference elsewhere either.
(Obviously making the fee gigantic would work, but good luck implementing a $1 bag fee or whatever...we already charge only 99 cents plus tax for our cheapest reusable bags, which people definitely do not reuse hundreds of times, because they're treated as trendy novelty collector's items as much as functional satchels. Never let a good mandate go to waste...)
*unless a customer is being an asshole, in which case we absolutely will, as all such small power moves go
I live in the UK, work in energy and infrastructure, and beg to differ! Wind is really great in certain areas, and offshore especially, whilst more expensive, can have surprisingly high load rates and stability.
> Anyone can suggest an additional alternative and pay a fee, for it to be considered, within a fixed time window.
This would be a major change, perhaps more than Zvi anticipated, because the current practice is to come up with a bunch of unrealistic alternatives, that you can reject, and not consider anything that would actually compete with the proposal. This is doubly true if the project is some politician's signature project (e.g., train to the airport) rather than something really technocratic.
I think Zvi has correctly identified the core disagreement here: "Any given paperwork should either have real value of information, telling you whether or not do something, or it shouldn’t be required."
This is in dramatic contrast with not only the purpose of NEPA, but how law/bureaucracies treat ...everything.
"Alex Trembath: When you ask people if they’re in favor of nuclear power *for generating electricity,* as opposed to just asking if they’re in favor of nuclear power, their approval jumps substantially."
I really wonder whether people know the difference between nuclear power and nuclear weapons. Or they read polls with only half their brains.
"Alec Stapp: Insane week for permitting reform (really feels like we've reached a tipping point):
1. NEPA exemption for chip fabs passed Congress
2. NEPA reform for wildfire prevention passed House
3. NEPA catex for geothermal passed House
4. And now Harris calls for permitting reform
It’s great that we’re making progress on all those bills and calling for reform. That there are so many is a sign we’re going about this in a not ideal way, but it’s far better than not doing anything."
I want to register disagreement with this (and the eliminating NEPA for public investments proposal). This is basically central planning by stealth; with political favor you get money and exemptions from the rules, without it you're screwed.
I also think the Republican Party deserves credit for being clearly superior on this class of issue (hence the IRA investment and cheaper housing). I don't think this will change, because there are both ideological (Republicans are much more pro-free market and pro-capitalism) and coalitional (the NGOs, regulators, and unions are overwhelmingly (D)) reasons for this. Balsa's intended to be nonpartisan, but it's hard not to notice the (D) lean of this article; many of the plaudits Democrats (AOC, Walz, Smith, Yglesias, Klein) in this piece are getting for saying the right things are things the Republicans have actually been doing for decades.
Not buying the pro-plastic bag gloss-over. Smells like serious cherry-picking. When I catch-up reading your posts I'll do a deep-dive on how they calculate plastic:cloth bag enviro ratio (but my first guess is bc cloth requires a lot of water in manufacture, and water is a bad pollutant? huh?).
As for 120 uses of a reusable bag? Watch me, no problem!
Do you think you are going to spread disease by re-using a bag? I do not thinkk you seriously believe that, but if you do, don't ever go outside, ever, again.
Love the big picture, tho, and for the record: Chevy Volt, I fly like once a decade (Amtrak!), and low-meat diet. Now, I did move out of a city (Rochester counts) but that was to reduce my commute to 4 miles.
On AI in the NEPA process: "Whereas the defenses and courts have limited resources and face asymmetric costs dealing with this. They are the ones producing thousands of pages of documentation, and who have to deal with every single complaint, and so on. So that suggests they are favored."
I work at the Department of Energy and this project is in my portfolio. It's part of a larger AI+permitting initiative. So far the PNNL team has built and continues to add to a large data lakehouse of NEPA and related federal permitting documents, using AI to extract features and metadata from the raw PDFs. They have built an AI-augmented search function on top of this dataset, and they have built a chat application layering LLMs on top of this dataset. In fact, they also fine-tuned an open source model that outperforms frontier models on their NEPA benchmark. DOE has a MOU in place with the CHIPS office for them to use this tool and we are working on getting MOUs in place with other agencies as well. Coming soon: AI-augmented public comment analysis. https://www.pnnl.gov/projects/policyai
Podcast episode for this post:
https://open.substack.com/pub/dwatvpodcast/p/nepa-permitting-and-energy-1
Link for list of environmental groups opposing permitting reform bill goes to the Bloomberg article on subsidies mentioned further down.
Really glad to have another series of this nature; I love me some roundups!
I don't think we're anywhere near maxing out the cost curve for solar. I think we may be close to maxing out the cost curve for traditional utility-scale installations, though. The cost of the panels themselves is a pretty small--maybe 20% of the final cost. The other cost is for things like the inverters, mounting hardware, labor, etc. But you can drive down the cost of thoses aspects quite a bit.
For example, you don't need an inverter. Lot's of applications, such as data centers, electrolyzers, or even HVDC transmission can operate on DC power. Another significant cost is mounting hardware, to keep the solar panels in an optimal position relative to the sun, either tracking or fixed. But you can also just drop the solar panels on the ground! You lose a bit of efficiency, but you can make up for that by using more solar panels. Oh, and this also cuts down by a huge amount on the labor needed to install the mounting hardware and mount the solar panels.
Some calculations on ChatGPT tells me that a single standard-sized container could hold about 10 MW of thin-film solar panels. You could conceivably manufacture a container-sized device that automatically unrolls 10 MW of solar power onto unused desert land with virtually zero mounting hardware, inverter, and labor costs, taking full advantage of economies of scale. There is a lot of space left on the learning curve to reach that point.
I truly feel that houses in the near future will have DC current as the default (except perhaps special feeds to kitchen, laundry room. water heater.)
It makes no sense to have tiny inverters in every lightbulb and computerized device.
very interesting...
Isn’t the reason for AC that it makes voltage transformation easy? If you trade your inverters for more complex transformers … you might end up with a win, but it’s not clear to me you do.
Alternatives run everything with the same voltage (not great) or have multiple circuits with different voltages (probably also bad).
Yes, that was the reason for AC. But DC to DC converters are much cheaper now due to better computer-controlled solid-state switches.
I can't for the life of me understand the Sun Cable proposal. I've worked in energy, but I'm not an engineer. Perhaps some smart folks can tell me I'm wrong, but besides some kind of "blue sky thinking" grift, how on earth does it make sense to link Darwin to Singapore with a power cable? The distance from DRW - SIN is greater than the distance from DRW - SYD. It's much greater than DRW - BNE, which would also be a more logical point to link Darwin to Australia's east coast electricity grid. Surely it's cheaper and less risky to build overland across Australia than laying undersea cable through (largely) Indonesian waters? Can it really work out cheaper laying deep sea cable to South East Asia rather than connecting Top End solar to other Australian locations? Is the power price differential, or demand level in Singapore, really large enough to justify such a cost and risk?
The only sensible part is the time difference - when regional Northern Australia is generating power at midday and afternoon, Singapore is coming online in the morning. Power going to East Coast Australia doesn't have this advantage I suppose. But still, midday power from Northern Australia would be selling into the evening peak on the East Coast.
It doesn't seem to make sense. It seems there is some bias in people's thinking where they assume that Singapore is close to Darwin, because they're all tropical, right? Now Australia's big, but so is the distance between Aus and South-East Asia.
Undersea data cables are actually considerably less expensive per mile than overland. I don't know if the same dynamic holds for power cables, but if so that would explain the apparent discrepancy in the plan.
Also, a lot less permitting and land ownership issues if you go by sea.
To follow on my earlier point: many fibre optic cables connecting countries in South America and Africa actually just bunny hop across the coast, because that's where the population is and it's easier to build!
It's because Singapore wants to buy way more power than the rest of Australia does; this is very much a one-way cable in practice. I also assume (but might be wrong) that power sold in Singapore can make its way into Malaysia too.
The big advantage of the Northern territory is a big sunny desert, but other Australian states also have big sunny deserts of their own. SEAsia has rainfall and thus cloud cover, and way less empty open flat land.
Re: banning plastic bags; the various UK administrations have mandated a (very small- I think it's 5p) charge on disposable plastic shopping bags. I've no idea what the effect is on plastic use, but it certainly has improved the littering situation; places just look nicer with fewer plastic bags around the place. I mean, yes, a ban is worse than a tax, sure. But I'm not sure that I oppose that move.
I do oppose the ban on plastic straws (we have that in the UK, too); that is pretty unarguably silly. But only technically; I don't really care about it personally.
The tiny fees are in theory fine, but my experience is that transaction costs are substantial. If it actively did reduce the number of bags, the question is whether it was due to the imposed transaction costs or the actual fee.
I'm not convinved such bag fees actually decrease usage, or even littering...though of course the locales aren't quite comparable. My store doesn't do plastic bags anymore (one-off experimental thing during some Christmases and covid, some customers love them, most hate them and they're super annoying for us cashiers to fill), only paper, and we're supposed to charge...25 cents(!) per bag. That is, 50 cents for a double bag, which is the most common usage cause everyone's paranoid about singles breaking. People on foodstamps are exempt by law - talk about cross-pressure incentives - but in practice...we mostly just don't bother assessing the fee* at all. It feels super petty, customers hate it, the revenue goes 100% to the government, there's no accountability for collection, it slows down checkout flow, and most importantly it doesn't actually deter people from treating the bags as free and deserved. The most price-sensitive people will still carry all their stuff bagless, or improvise (free) produce bags as shopping bags, so it helps slightly on the margin. But these thrifters are not responsible for the bulk of bag consumption anyway! That's the families of 4 who spend hundreds of dollars and end up with dozens of bags per cart, as well as the buy-a-single-item kids who somehow need a bag for their bag of chips. Both are insensitive to the fee, the former because it's a rounding error, and the latter because they're almost always paying with Someone Else's Money so who cares. If such a mandate can't work in SF, I don't expect it to make a difference elsewhere either.
(Obviously making the fee gigantic would work, but good luck implementing a $1 bag fee or whatever...we already charge only 99 cents plus tax for our cheapest reusable bags, which people definitely do not reuse hundreds of times, because they're treated as trendy novelty collector's items as much as functional satchels. Never let a good mandate go to waste...)
*unless a customer is being an asshole, in which case we absolutely will, as all such small power moves go
100% agree on permitting and nuclear. If anyone wants a deep dive into why nuclear is safe, and why it is expensive, this podcast is the place to go: https://humanprogress.org/robert-zubrin-nuclear-energy-space-and-humanitys-future/
Partially agree on solar and geothermal, and hard disagree on wind and especially offshore wind. Windpower is not the future.
I live in the UK, work in energy and infrastructure, and beg to differ! Wind is really great in certain areas, and offshore especially, whilst more expensive, can have surprisingly high load rates and stability.
> Anyone can suggest an additional alternative and pay a fee, for it to be considered, within a fixed time window.
This would be a major change, perhaps more than Zvi anticipated, because the current practice is to come up with a bunch of unrealistic alternatives, that you can reject, and not consider anything that would actually compete with the proposal. This is doubly true if the project is some politician's signature project (e.g., train to the airport) rather than something really technocratic.
I think Zvi has correctly identified the core disagreement here: "Any given paperwork should either have real value of information, telling you whether or not do something, or it shouldn’t be required."
This is in dramatic contrast with not only the purpose of NEPA, but how law/bureaucracies treat ...everything.
"Alex Trembath: When you ask people if they’re in favor of nuclear power *for generating electricity,* as opposed to just asking if they’re in favor of nuclear power, their approval jumps substantially."
I really wonder whether people know the difference between nuclear power and nuclear weapons. Or they read polls with only half their brains.
"Alec Stapp: Insane week for permitting reform (really feels like we've reached a tipping point):
1. NEPA exemption for chip fabs passed Congress
2. NEPA reform for wildfire prevention passed House
3. NEPA catex for geothermal passed House
4. And now Harris calls for permitting reform
It’s great that we’re making progress on all those bills and calling for reform. That there are so many is a sign we’re going about this in a not ideal way, but it’s far better than not doing anything."
I want to register disagreement with this (and the eliminating NEPA for public investments proposal). This is basically central planning by stealth; with political favor you get money and exemptions from the rules, without it you're screwed.
I also think the Republican Party deserves credit for being clearly superior on this class of issue (hence the IRA investment and cheaper housing). I don't think this will change, because there are both ideological (Republicans are much more pro-free market and pro-capitalism) and coalitional (the NGOs, regulators, and unions are overwhelmingly (D)) reasons for this. Balsa's intended to be nonpartisan, but it's hard not to notice the (D) lean of this article; many of the plaudits Democrats (AOC, Walz, Smith, Yglesias, Klein) in this piece are getting for saying the right things are things the Republicans have actually been doing for decades.
Not buying the pro-plastic bag gloss-over. Smells like serious cherry-picking. When I catch-up reading your posts I'll do a deep-dive on how they calculate plastic:cloth bag enviro ratio (but my first guess is bc cloth requires a lot of water in manufacture, and water is a bad pollutant? huh?).
As for 120 uses of a reusable bag? Watch me, no problem!
Do you think you are going to spread disease by re-using a bag? I do not thinkk you seriously believe that, but if you do, don't ever go outside, ever, again.
Love the big picture, tho, and for the record: Chevy Volt, I fly like once a decade (Amtrak!), and low-meat diet. Now, I did move out of a city (Rochester counts) but that was to reduce my commute to 4 miles.
agree but its important to note how fast solar has been growing across the US, for example PJM has seen like 70% YoY growth in max solar output:
https://www.gridstatus.io/records/pjm?record=Maximum%20Solar. So even despite all these regulatory challenges, something is working right, with no sign of slowing down anytime soon
On AI in the NEPA process: "Whereas the defenses and courts have limited resources and face asymmetric costs dealing with this. They are the ones producing thousands of pages of documentation, and who have to deal with every single complaint, and so on. So that suggests they are favored."
I work at the Department of Energy and this project is in my portfolio. It's part of a larger AI+permitting initiative. So far the PNNL team has built and continues to add to a large data lakehouse of NEPA and related federal permitting documents, using AI to extract features and metadata from the raw PDFs. They have built an AI-augmented search function on top of this dataset, and they have built a chat application layering LLMs on top of this dataset. In fact, they also fine-tuned an open source model that outperforms frontier models on their NEPA benchmark. DOE has a MOU in place with the CHIPS office for them to use this tool and we are working on getting MOUs in place with other agencies as well. Coming soon: AI-augmented public comment analysis. https://www.pnnl.gov/projects/policyai