Why I Hate S-Curves: An Interview with Rob West from Thunder Said Energy
This week, our guest is Rob West, founder and CEO of Thunder Said Energy. Founded in 2019, the research firm aims to help decision-makers find economic opportunities in the energy transition. Rob lives nine time zones away in Estonia and is an uber-productive energy expert who covers a wide range of topics in his consultancy.
Rob explains why he hates S-curves, since they are often used to assume future growth rates for new energy technologies instead of properly analyzing what is realistic.
Here are some of the questions Peter and Jackie asked Rob: What do you think of the recent sell-off of clean energy stocks and the reasons behind some company failures? How much do you think the electricity demand will grow due to AI? Will there be investment in new natural gas-fired generation to meet this demand growth, despite the concerns about carbon emissions? Why do you think solar is the most important energy source in the world? Why do you see power grids as the biggest bottleneck in the energy transition? Do you think naturally found hydrogen in the earth’s subsurface, also called “gold hydrogen,” could be the next “gold rush”?
Other content referenced in this podcast:
- Thunder Said Energy website, see past research notes and sign up for daily email: https://thundersaidenergy.com/
- FT article “More cleantech companies fail as fundraising challenges emerge (ft.com),” September 2, 2024
Please review our disclaimer at: https://www.arcenergyinstitute.com/disclaimer/
Check us out on social media:
X (Twitter): @arcenergyinst
LinkedIn: @ARC Energy Research Institute
Subscribe to ARC Energy Ideas Podcast
Apple Podcasts
Amazon Music
Spotify
Episode 251 transcript
Speaker 1:
The information and opinions presented in this Arc Energy Ideas podcast are provided for informational purposes only and are subject to the disclaimer link in the show notes.
Speaker 2:
This is the Arc Energy Ideas podcast with Peter Tertzakian and Jackie Forrest exploring trends that influence the energy business.
Jackie Forrest:
Welcome to the Arc Energy Ideas podcast. I’m Jackie Forrest.
Peter Tertzakian:
And I’m Peter Tertzakian. Welcome back. It is now September. I flipped my calendar, although today digitally that just means swiping, but it is September. We had the September long weekend. What’d you get up to?
Jackie Forrest:
I did some fun things. One thing I did is, I don’t know if you know about this, Peter, but they opened Highway 1A between Banff and Johnson Canyon, well, they close it to cars, therefore bikes are on this highway with nobody else and it’s so beautiful.
Peter Tertzakian:
Oh, yeah.
Jackie Forrest:
So I did that.
Peter Tertzakian:
That’s gorgeous. Yeah, I’ve done that several times.
Jackie Forrest:
Yeah, it’s 40 K return. I actually did it with Molly in a cart behind me.
Peter Tertzakian:
Your dog? Yeah, yeah.
Jackie Forrest:
Yeah, but Molly was not a big fan of being in this cart.
Peter Tertzakian:
Isn’t she bait for bears?
Jackie Forrest:
Yeah, I guess so. But luckily we didn’t see any bears. But yeah, I won’t do that again. I won’t take her in that cart for 40 kilometers. But it was beautiful. And the e-bike, first time I ever did it with an e-bike, a lot more fun than the other way.
Peter Tertzakian:
I’ve done it with both. I’ve it with a regular road bike and an e-bike. Yeah, there’s merits to both. There’s a couple of the hills are fairly taxing on a regular bike. But no, it’s good. It’s fun.
Jackie Forrest:
Yeah, so we still got a nice fall. That’s actually open for the whole month of September, I understand.
Peter Tertzakian:
It’s going to be gorgeous. We’ve got a hot couple of days coming up and the weekend prior to us recording this looks just great. Okay, well, lots to talk about. We want to bring on our special guest, one of our all-time favorite guests from the past.
We’ll keep you in suspense here for a couple of minutes, but we want to talk about this Financial Times, FT, article from September 2nd, more Clean Tech Companies Fail as Fundraising Challenges Emerge. And we bring that up because we talked extensively about the challenges of financing energy at large, including clean tech, last week’s podcast, and I think this sort of validated a lot of the things we’re talking about. As of time of recording, once again, a lot of the tech type stocks, clean tech in particular, the stock prices, share prices have fallen, which means the cost of capital has risen. And I’ve always said no money equals no transition, because if you can’t finance these companies beyond the public purse, in other words government incentives, if we don’t have private investment capital following through in multiples of public financing, then the transition is going to be very challenging.
Jackie Forrest:
They also had in the article a number of really high-profile investors that have been supporting clean tech like Amazon, TotalEnergies, Bill Gates have been involved in companies that they cited as going bankrupt because they raised a bunch of money two to three years ago, ran out of money, and there was no ability to raise additional capital.
Peter Tertzakian:
Yeah, the ability to raise money two to three years ago at the start of the Biden Administration, so actually it’s pushing four years ago, was easy. I mean, it was the big surge in interest in clean tech stocks, made it very easy for these companies to raise money. But when the share prices fall, all of a sudden they’re locked out, and if they have not had their product adopted and they don’t have sufficient cash flow to keep going and there’s limited sources of raising money, then it becomes very challenging. And that’s where we’re at. So we’ve got bankruptcies, the article mentioned battery companies and other companies. So I think we want to talk about that some more and we want to talk about it more, even, with our special guest who’s about to come. So who do we have, Jackie?
Jackie Forrest:
We want to welcome Rob West, analyst and CEO at Thunder Said Energy and he’s a repeat guest. Our last conversation was about a year ago from October 2023. We had great feedback, Rob, so thank you so much for coming back on the show
Peter Tertzakian:
Yeah, all the way from Estonia.
Rob West:
Yeah. Well, thank you so much. Great to be with you.
Peter Tertzakian:
Good. As to remind our audience that you actually live and work in Estonia, but maybe, Rob, remind them of Thunder Said Energy and what you do.
Rob West:
Yeah, I wanted to find somewhere that’s colder than Canada. That’s part of the appeal. The focus really for me is just we want to get the world to net-zero. We want to meet the energy needs of the world and take out all the CO2 while making energy better, cooler in the process, and what are the options? And just start from first principles. For every single thing, what is it? How does it work? What does it cost? Why does it cost that? What are the challenges and who’s doing it?
I think your question about failing clean tech companies, I’m going to put a positive spin on it. The positive spin is it’s actually it’s okay to fail. Failing is, in a sense, inevitability of trying new stuff. I think if I was going to have a somewhat salty complaint it’s that we could fail better. I would say that some of the new energy concepts that might be on the fail list, if you have a modicum of thermodynamic analysis or cost analysis or some of the notes I guess we’ve been trying to write, you might have looked at those things and some of them said, “Well, that’s the thermodynamic equivalent of printing out your emails. I’m not sure that one warrants the attention.” But there are also some great options. And inevitably you move from this phase of, “This is cool, we should explore this,” to the gritty reality of, “This is hard, we can’t make this work.” And that’s going to happen if you throw a lot of darts at the dartboard.
Peter Tertzakian:
I completely agree with you. Failing is a part of the innovation cycle and certainly in capitalism as well, and that we have seen this before, all the way from Edison’s era and the light bulb companies and the electrification companies all the way to today. So I certainly wouldn’t debate you on that. But I think your comment about, I think what you’re alluding to is the potential inefficiency of the allocation of capital due to excessive hype and those who would argue that we don’t have the luxury of misallocating capital and failing excessively, that if the funds had been channeled a little bit better, that we would be further ahead today. What would you say to that?
Rob West:
I do take that point. If the ultimate investment needed for energy transition in my roadmap is about $9 trillion a year and we whiff on 10% of that, here’s a sort of a trillion dollars a year of misallocation, which is quite a lot to misallocate. But I think there’s a couple of things going on here and no doubt it has been a really, really bad year for the S-curve enthusiasts, and maybe we’d come back to that and why I hate S-curves.
But I actually think the bigger thing of what’s changed is just attention is going elsewhere. I’ve been looking at this. Amazingly, in the upcoming US election, climate and decarbonization is really only a critical topic for maybe 15% of the electorate. It’s certainly way behind cost of living or apparently also border issues. And I think there’s something in that. I think another trillion dollars a year is going to be spent on defense by 2030, which probably wasn’t in people’s expectations prior to 2022. And other things are just on the agenda. There’s potentially catastrophic bottlenecks in our power grids, there’s rising volatility in the energy system, there’s the whole AI theme, which has just been this rabbit hole where the more I have written, the more people that seem to want me to write. So I think that has stolen some of the thunder from things that might’ve been really exciting three to five years ago.
Jackie Forrest:
In fact, that article talked about the fact one of the issues is there’s so much excitement around AI and some of the companies around that it’s starving the clean energy sector of capital and also making people think, “Well, growth doesn’t look that high there,” relative to some of these other sectors.
Peter Tertzakian:
But I think it also points to the clean energy space being oversold and overhyped, and that comes back to bite you. Because it was positioned as being rapid adoption just like what you see in the information technology space, including AI. “Get on board, you’re going to see the rapid substitution of fossil fuels. This is going to be incredible.”
In my mind, it’s not so much that there’s AI and other things as alternatives that have faster adoption, it’s actually the fact that the space was oversold. And you prove that in your many wise writings about the technological challenges, the uneconomic nature of many of these pathways, or I would just say uneconomic today, I think we can put a positive spin on what may happen tomorrow. We’ll talk about that as well. But would you concur that the root issue was the space being oversold?
Rob West:
I don’t know. I think the thing that has depressed me the most is that some of the things that make the most thermodynamic sense and should over time make the most economic sense have been some of the things that have been hit hardest. So in a way it comes back to my hatred of S-curves. Like, Peter, your point I think is exactly right. You can look at these technologies like refrigerators or televisions that basically came in, either in the 1920s or 1940s, and then within 15 to 20 years had a hundred percent market penetration. Everybody had one. There’s this temptation when you see EV sales going from 2% of all the vehicles in the world five years ago that are sold to 15% of all the vehicles sold in the world last year, to say, “Oh, yeah, that’s an S-curve and give it 10 years and it’ll be a hundred percent of all the vehicles getting sold.”
But when you sort of break it down according to the demographics, amazing stat, 80% of all the electric vehicles been bought by the upper 30% of earners. Another breakdown, 60% have been bought by the upper 5-7% of earners depending on how you quantify. And then it gets amazing. The average EV buyer has an income over a $100,000. And when you sort of plot the demographics, you can make a good argument that we are now saturated in terms of we could just roll forward at 15% of sales between now and 2030. And it’s kind of rare in your career as energy analyst, or as, I guess, everything analyst, I fear that you sort of think of something like that and you think, “God, if that’s true, then I can think of 50 names that are the cells.” I don’t think it will be quite as bad as a plateau at 15%, but something in the range of 15 to 30 feels right to me out to 2030 for EV sales. And similar for solar. Last year, a great number of modules sold, but by liquidating a lot of inventory, like real bargain-basement prices, and now if you look at it in the U.S., solar growth is more than electricity growth. And that’s interesting. It’s not the first time we get to that point, but what that means, if you think about it, it’s not like, “Hey, we’ve got demand growth for electricity. What should we add to meet that demand growth?” If we wanted to grow solar more from here, it would actually have to displace something that’s already there, already built, already running. And so the economics become very different.
Peter Tertzakian:
Right.
Rob West:
And so lots of different reasons for slowdowns, but this idea of exponential growth forever I think has had a bit of a year of reckoning.
Peter Tertzakian:
It has had a year of reckoning. I want to get back to the S-curve notion, and for our audience, the S-curve is when a new product that is compelling comes in, it takes a while, slow start to get adopted, and then there’s an inflection point after about 3% penetration of the market. All of a sudden, you tend to see really rapid sales and adoption almost becoming not vertical, but the very, very steep adoption rate. And then as the market gets saturated, it starts to mature and flatten out so the whole shape looks like an S. And I think, Rob, what you’re saying is, in the demographic sense, we’ve had an S-curve penetration of electric vehicles. In other words, the upper income people and early adopters helped steer an S-curve. We plateaued out.
But I’m personally a believer actually that we are going to see another technological breakthrough in batteries probably in the next five years. The costs are coming down. Actually, the Chinese are proving that with their low cost vehicles. And then we’re going to start to see the S-curve built in the broader populace.
Jackie Forrest:
Let’s move on from the hate of S-curves, but I get your point in that basically people have put in these very simple models that it’s all going to follow this very generic S-curve and we’re finding complications, and generally things are going slower. But I do want to talk about, you also mentioned that one of the reasons is there’s a new shiny ball that is getting investors’ attention, which is AI and electricity demand. And since we last talked, you’ve done a lot of work in this area. So can you just tell us, how much do you think AI will add to electricity demand and how much uncertainty is there in predicting that?
Rob West:
Sure. So I think we’re going to have 150 gigawatts of AI data centers globally by 2030. I think that about 60 gigawatts, so 40%, is going to be in the U.S. and maybe broader North America around the fringes. I think that’s going to be about 1,000 terra watt hours of useful electricity consumption by 2030. And my attempts to do a confidence interval is somewhere between 300 and 3,000 terra watt hours.
Peter Tertzakian:
Just to put this in perspective for our audience, isn’t it like one gigawatt is roughly one nuclear power plant?
Rob West:
Yeah, yeah, yeah, that’s exactly right.
Peter Tertzakian:
You have to build 150 nuclear power plant equivalents by 2030 in five years.
Rob West:
The other one I was going to give is that, I seem to be unable to leave this topic, but my total electricity consumption for electric vehicles in 2030 is about 500 terra watt hours. So all of the energy consumption of AI is going to be about twice as much as all of the electric vehicles in the world in 2030.
Peter Tertzakian:
Crazy.
Rob West:
Big, big numbers. And so it really changes things because, if you look everywhere, electricity demand is inflecting. Globally, we’ve added, total electricity demand is about 30,000 terra watt hours per year, it’s been growing at about 600 terra watt hours per year and it’s going to grow at about 1,500 terra watt hours per year going forward. So if you look at the U.S., flat for the last 20 years and now we’re looking at 2.5% growth. Europe, down 0.6. We’re looking at 1% growth. Japan, same story. Every market, you’re having that.
Peter Tertzakian:
And this is crazy, Rob, because what we talked about on our podcast last week, Jackie, was that we’re having difficulty now financing any kind of energy.
Rob West:
Right.
Peter Tertzakian:
We talked about the cost of capital rising for the clean energy space and we talked about the aversion to financing fossil fuels including natural gas-fired power, which can be brought on quickly to meet this demand. Nuclear power has its issues including high capital costs and lag time to get them built rather than potentially some of these modular reactors. But even so, they take three to five years. And here we are tripling the growth rate of electricity. Does that paint an unsettling picture for you?
Rob West:
Well, I mean, I was thinking about that and how it’s happened in Europe. I did this note looking at energy security. Punch line, 70% of the energy consumed in major economies is produced domestically and 30% is imported. And actually, if you stack up all the regions, Europe has been the worst. In the last 30 years, we’ve gone from 65% self-reliance to about 40% self-reliance despite ramping renewables to 10% of our mix, which is all domestic. And the reason is that we pulled back on nuclear, on gas, on all self-generated energy except the very cleanest of the clean when really the right answer is to be pragmatic and build anything that’s a seven out of 10, should really should get momentum behind it.
So I think that where that leaves the demand that is coming is I do think over time it is going to result in more energy being developed and consumed because our old friend, the first law of thermodynamics, you can’t consume what you don’t have. But I think in the near term, the cascade is as follows. AI data centers are just going to buy generation that already exists. They actually want a very specific set of things. They want clean power that is available now, and those are the top priorities. And so great, buy all the output of existing nuclear plants, existing hydro plants, and you can have it tomorrow. You just have to find the right strike price.
Peter Tertzakian:
But how will society balance AI versus potential brownouts?
Rob West:
Well, so I don’t think it causes brownouts. What I think is it feeds into how electricity markets work and it’s going to require increasing capacity payments and increasing construction of other generation sources. So here’s how it works. So if you think about every region, and there’s a few 100 across North America, they have a regional grid planning authority who is in charge of maintaining a reserve margin. And the regulations is run by NERC, the North American Electricity Reliability Council. It includes the U.S. and Canada. It basically says you must have no more than one year in 10 where you could have a loss of load event or an LOLE. So that’s meant to happen less than one year in 10, and to do that you have to maintain at least a 15% reserve margin.
And so data centers come along, they’re like a wolf at the door. They look for these spare capacity chickens and they want to gobble them all up. And that’s fine. We have a free market for capacity across North America. If somebody’s willing to pay for the capacity, they can pay for it, they can buy it. That’s how it works. But what it means is that, if I gobble up your spare capacity chickens and I bring them all over here, your grid now doesn’t have enough spare capacity and, as a result, you as the grid operator have to go and finance or tender for people to build more capacity somewhere else in the system. And I think about two-thirds of that ends up being CCGTs and peaker plants. And that’s how I think the pig feeds through the snake.
Jackie Forrest:
By the way, we should mention, we will put a link, Rob, if you don’t know, has these great daily notes and some of the research we’re going to be referring to is in these notes. So I will put a link for you to sign up to his daily notes, and we also signed up to getting the more detailed research. But I want to talk about some of your writing has made me quite bullish on natural gas because it seems to be the only thing that can meet this demand maybe. Am I right on that? And do you think natural gas is going to meet this need, especially in a world where we’re very concerned about emissions and there’s real concerns about the longevity of these brand new natural gas plants. Are they going to be allowed to operate for 20 years? I mean, here in Canada there’s policy being drafted that’s saying after 20 years you won’t be able to run them or you’re going to have to put carbon capture storage on them. And are we going to get people wanting to invest?
Rob West:
Well, I think that’s okay. That’s what makes a market. What makes a market is that you’re going to have differences of opinions on what assets are worth. And actually, in a way, this is what creates opportunity for people is, when there’s this insane polarization about, do we need this thing or not? It’s where valuations change. So I think that if you look at deal prices for natural gas turbines over the last decade, they average about $500 a kilowatt. And recently they’ve been running about $1,000 a kilowatt, and I think they’re worth $1,300 a kilowatt or more. And the big thing that’s doing that, actually, it’s volatility. We all love really simple models of generation assets where you say, what’s the levelized cost, which is a made up number that means nothing because what’s relevant is the cost you can actually access.
And so if you line up the distribution of power prices throughout a year, it’s not the same number every one of the 8,760 hours. It’s a log normal distribution. I don’t want to scare people to just switching off because I’ve talked about statistics, but what that means is, it’s an asymmetric distribution. The tail to the upside, it’s like when power prices are high, they’re really, really high, and when they’re average, they’re average. And so what has happened in our power grids in the last 10 years is that the volatility has traveled. That’s because of renewables getting built in. Renewables are volatile. Over about 500 square kilometers, the output of wind and solar are between 60% and 90% inter-correlated with themselves. And so either it’s sunny or it’s not, or it’s windy or it’s not, and in those moments when grid demand is high and the wind isn’t blowing and the sun isn’t shining, power prices have been going bananas. And you can really see that in the data. Actually, if you look at somewhere like ERCOT, median power prices are exactly the same as they were 10 years ago, but mean power prices are up 60% and it’s all in that upper tail.
Peter Tertzakian:
That’s in Texas, right? ERCOT? Yeah. Yeah.
Rob West:
Yeah. It’s like across all the developed world grid that I’ve looked at. What power prices do gas turbines access, they specifically ramp up at the upper end of that distribution. They hit the peak prices, not the average, but the peak. And that’s why I think they’re worth more is that yes, you might have these peakers that only run between two and 30% of the time, but the prices that those peakers actually access have been going up because that curve is getting steeper. That upside tail has been growing larger. That’s why it’s interesting.
Jackie Forrest:
So even though you may think, oh, this is an asset that I can only make money off for the next 15 or 20 years, you actually can get a payback because of that volatility. So therefore people may still invest.
Rob West:
Well, I don’t think they’ll ever phase them out. I have looked at, I want to help my clients invest in batteries, but I’ve really struggled. You need 20 cents a kilowatt-hour as a sort of daily round trip storage spread to start making a 10% return. And if we ever get there for batteries, I’m telling you gas turbines will be printing money in the four to five years before we get there on the batteries. And that’s kind of why it’s interesting for me.
Peter Tertzakian:
See, I mean, the thing is that if gas turbines are printing money, and you already talked about the mean price of electricity going up in markets, I mean this now relates to the number one issue in the minds of people and that is affordability. And so you’re painting a picture here of the need for 150 equivalent nuclear power plants because we’re starting to do all these AI prompts and getting AI to make all these silly images, which take just an inordinate amount of computing power and therefore energy. And so maybe I was too dramatic earlier saying, would you rather have AI or brownouts? Maybe it’s the question, would you rather have AI or high electricity prices? And you’re arguing that the market, okay, will decide. But what we’ve seen in the last 10 years in particular is that the government intervenes aggressively in energy markets, whether it’s for sustainability purposes, climate purposes, and now affordability. I think these markets are not going to operate in a free market sense, and we’re going to see governments trying to play a whack-a-mole game trying to figure out how to deal with this extremely complex problem.
Rob West:
I like your concept of energy whack-a-mole. I think if I was going to rename my business, that might be a good candidate, but I’m more optimistic than that. I think that most of the energy demands of AI will not be artificially generating cat videos. I think it’s going to be useful stuff. There’s a great stat, which is the average industrial process is 20% efficient. And for 25% of the industrial processes, they’re less than 5% efficient. And I love that example of aluminium and how before the whole Heroult process aluminium was impossible to smelt. It was a 2% efficient process. Aluminium was so rare that Napoleon had his cutlery made out of it.
And now we sort of have the whole Heroult process. It’s like 50% plus efficient. We wrap our sandwiches in the stuff, it’s available everywhere. And it sort of makes you think, well what about copper? Why couldn’t we find the ultimate bio leaching process to extract copper from chalcopyrites? Or why couldn’t we take vehicles? And you now have autonomous vehicles driven by AI that are going from one end of a city to another and driving 20% more efficiently than any human driver can.
Peter Tertzakian:
Well, we are going to see that. And I think AI holds great promise to do those sorts of things and make our systems more efficient. But you know better than anyone, the whole Jevons principle that if you make it more efficient, people just buy more aluminium, people just want more cars.
Rob West:
You can’t throw Jevons back at me, but you’re right. You’re right.
Peter Tertzakian:
Right. The more efficient you make something, then the more people demand of it and,-
Jackie Forrest:
Okay, I got to get this point in though.
Peter Tertzakian:
Okay.
Jackie Forrest:
So do you remember when David Hobbs came on and we had this debate and you guys made fun of me for not knowing enough about Star Trek and they were basically saying, do you think the people in the Star Trek world use less energy or more energy than us today and that people will continue to use more energy? But Bill Gates recently said that while AI will increase power demand in the short term, in the long term, it will identify areas to reduce energy use. So Bill Gates agrees with me and it sounds like Rob does too, by the way.
Peter Tertzakian:
I don’t agree. I don’t agree at all. I think when you make processes more efficient and better people just demand more of the products that that process produces, and I think that humankind has a voracious appetite for consumption.
Rob West:
So I think both can be true. And that’s really exciting, is that we can have more consumption. Yay. I like consuming stuff. Now we can have more efficient consumption. Yay. I like efficiency. And as an investor, this is where you want to go, is you want to find ways of doing stuff better, more efficiently and then open possibilities for people that didn’t exist before.
Peter Tertzakian:
So what are your views then on the sustainability issues, whether its more consumption leads to more pressure on everything from biodiversity to climate to all the environmental issues? I mean, we’re butting up against those big time.
Rob West:
Yeah, I think that is inevitability of having a planet of 8 billion people that will be nine and a half billion people. And that’s really what the entire point of energy transition is, is that we can meet the energy needs of human civilization in a way that doesn’t turn our planet into a wasteland. And I think if I think about the sort of hatred of S-curves and some of the turbulence we’ve had over the last year, part of the turbulence is the reckoning process where out of the woodwork come the really interesting things that we want to do. We want to do electrification, efficiency gains, big ramp up of solar energy, nature-based solutions, sensible amount of carbon capture. These would all be great things to do for energy systems and planet. And yeah, out of the turbulence, the real opportunities will come clear.
Jackie Forrest:
Talking about the opportunities for more sustainable energy, you sound very bullish on natural gas generation, but you’ve also said that solar is the most important new energy source in the world. So explain that in light of your view that we’re going to see a lot of natural gas generation come along.
Rob West:
Well, I mean if I generally have an observation for the last five years, it’s that I’ve written note after note after note, it goes like, this is cool. This is at the biting point of being real. I’m going to put it on an S-curve. And then I look back at those notes two to three years later and generally technology is hard. Most of those things have not worked and it makes the things that have worked and have broken through kind of all the more remarkable. And if I have a sort of observation, it’s that physical chemical process technologies. I mean, basically they take 15 years to go from the lab to the real world. And every time I want to write a note saying this physical chemical process technology will be different, I should stop and pinch myself and not publish that note. But semiconductor technologies have been remarkable, talking about AI. That’s one of them. Solar is a semiconductor technology. Actually, EVs are a semiconductor technology as well. But we’ll park that and we’ll go to solar.
20 years ago they were 5% efficient. 10 years ago they were 15% efficient. Today the best modules are 25% efficient. And if you look at what LONGi is doing, I recently went through their patents, we’re going to get multi-junction solar on the market in the next couple of years, and then we’re opening the door to 45, 50% efficient solar. And that’s cool. That’s going to change the world. It’s not to say everything will be solar. It’s not to say that we should do solar to the detriment of not doing anything else, but I think when kind of the turbulence shakes out, we are going to end up with energy generated by semiconductors. And that’s going to be a really big thing.
Peter Tertzakian:
I mean, I agree with you on the solar side. It’s a semiconductor and it has a very aggressive adoption curve. I won’t say S-curve, I’ll call it an adoption curve because the price plummeted last decade, especially during a period of low interest rates, which allowed the technology to be rapidly scaled and adopted. And it’s just absolutely amazing and I agree with you. But that solar is just very upstream and the mistake that’s been made in the rest of the clean tech space coming full circle to the beginning of our conversation is the expectation that all other energy conversion processes leading to end-use work and consumption is going to follow that solar pattern.
Rob West:
Oh, yeah. Totally, totally agree with that.
Peter Tertzakian:
That was sort of the mistake even in the financial markets which led to the bubble and now the collapse of the prices. Would that be,-
Rob West:
Not everything is solar.
Peter Tertzakian:
Yeah, not everything is solar.
Rob West:
I’m sitting in my office. I can see lots of things that aren’t solar, and it would be a mistake for me to think my coffee cup was solar or anything. It’s not. Right. And so I think getting the different models to apply to different growth curves is really important and good theme. The thing we haven’t talked about at all today, and one that I think we should is grid bottlenecks. Can we dive into that? I mean, it definitely relates to the solar discussion because that is also one of the things that maybe is slowing down wind and solar this year is we’ve gone from a world where these things used to connect up to the grid for about $25 a kilowatt within one to two years, and now it’s $125 a kilowatt. It’s gone up five times and it takes four to six years. That is how long the queue has become,-
Peter Tertzakian:
To plug and play.
Jackie Forrest:
Yeah.
Peter Tertzakian:
Yeah.
Jackie Forrest:
Well, let’s talk about that. You’ve done quite a bit about grid bottlenecks and you talk about power grids will be the biggest bottleneck in the energy transition, and we talk about the long lead time for transformers, and you’ve done work on that. We haven’t really seen any results and we haven’t seen brownouts. We haven’t seen problems yet.
Rob West:
But you won’t. You won’t. I mean, two things can be true at the same time, which is that we can have bottlenecks in our power grid, which are very real and very meaningful, and the lights can still be on. You might not know it when you go to the light switch, but they’re there. Then the sort of flashing red lights in the control room of hypothetical power grid operator are there saying, “Oh yeah, we have issues.” I mean, the biggest one is the long term one. You’ve been making this great point about the need to attract investment. Nowhere do we need to attract more investment in the energy transition than power grids. We did go from about $300 billion a year to about $1.5 trillion a year. And why? And here’s why. A lot of what we want to do adds electrons. It adds demand for electricity and the stat globally-
Peter Tertzakian:
Like cat videos.
Rob West:
Especially cat videos, that’s the number one. But every terawatt hour of electricity that you use adds about 225 kilometers of transmission lines that you need to support it and about 4,000 kilometers of distribution lines that you need to support it. Just the average rule across all countries in the world.
Peter Tertzakian:
Isn’t this the point? I mean, not only do we have to add 150 nuclear reactors that they have to plug and play by 2030 into this grid.
Rob West:
Or you have to find ways of not. So I’ve been looking at this for a year and I come to conclusion it’s impossible. You’ll never build this stuff.
Peter Tertzakian:
No, you won’t.
Rob West:
There’s a power line they’re trying to build in Germany that had to get 13,000 separate permits and it’s taken 15 years and that’s one of the good ones. And so it’s really, really tough to build this infrastructure anew. And so I think what’s been exciting me is what is the opportunity to take the existing infrastructure and upgrade it and get more power through it? And it can be raising voltage, it can be raising current, and it can get you into some absolutely fascinating value chains like reconductoring power lines or just the magic of superconducting cables where something the width of a human hair can carry the energy consumption of a country. And that’s cool.
Peter Tertzakian:
So this is akin to the situation with bandwidth, broadband for communications. There was always this talk about, okay, there’s just so many videos and other things competing for the very limited bandwidth on the copper cables and then fiber. But we’ve always been able to figure out how to push more through.
Rob West:
Yes, it won’t be as extreme as you’ve been able to do with fiber optics-
Peter Tertzakian:
Fiber optics.
Rob West:
But it will be the same. I think you can get maybe 20 to 50% more electricity through the same lines. That’s very real.
Jackie Forrest:
Rob, you did a really good paper on the potential for superconductors. The issue with superconductors is they need the cooling, so they’re very expensive. There is a limit, I think you can do it, but there’s going to be costs associated with it. Is it going to be economic to do things like superconductors, which would be really… How would that affect today’s copper wires if you could switch it to superconductors in terms of capacity?
Rob West:
Oh yeah. I mean, so you’re right, they are quite cold. They do have to be minus 200 degrees Centigrade. And so they are also about six times more expensive meter for meter. But the point is not to take transmission lines and make them superconductors. That is for very specific applications, mainly in cities. So if you think about between now and the 2040s, the average electricity consumption of the average house is going to double just roughly. You go from not electric car to something that needs to be charged. You go from not electric heating to potentially a heat pump. And then all these magical devices to play our AI generated cat videos. We’re going to have a lot of extra electricity demand per house.
And so if you think about a city, how are you going to tear up an existing city and get in what are some of the most expensive real estate in the world, just like a spare right of way to run more cables. And that’s where you’re never going to do it. And instead you can take the existing line and you can actually save land, you can actually take out land and substations that you no longer need because you can serve basically entire city off a few cables that carry the power through it. So it’s very limited context in the distribution network and it’s just another tool in the toolkit.
Jackie Forrest:
And I can see the economics working there. The land is so expensive that it’s worth it.
Well Rob, there’s so much to talk about and you’ve done such great research. We haven’t hit on everything, but I did want to ask you recently, you published a pretty detailed research paper on what natural hydrogen, which sometimes is called gold hydrogen. And some people are calling it this, in fact, I think you actually called it the possible a gold hydrogen rush is brewing. And I was listening to another podcast with a quote from one of the CEOs of, I think you talked about there’s like maybe 50 companies looking for this natural hydrogen, which is basically finding oil and gas, but now we’re going to find hydrogen just in the ground. So obviously it’s going to be a lot cheaper than using wind and solar to create hydrogen or converting natural gas into hydrogen.
This is a quote from the Koloma CEO Pete Johnson saying that gold hydrogen should be mentioned in the same breath as fusion energy in that it could be world changing for the energy transition. Now, I was reading your paper and I think there’s only one actual example of this in the world today where they’ve actually found this gold hydrogen and they’re using it for energy. So after you did all your research, do you think this should be mentioned in the same breath as fusion energy as a disruptive technology?
Rob West:
Oh god, Jackie, there’s no way I can answer that without getting in trouble. I think it would be really cool if we could find hydrogen in the subsurface and take it out of the ground. But sometimes research has a look over here approach and you sort of look and you think, well, for various reasons, personally, I find this one harder to get excited about. I think it will end up probably being more expensive than SMR hydrogen or turquoise hydrogen. And I think there might only be one major hydrogen field in the world, the infamous Bourakebougou hydrogen field in Mali. And it might be only one for a very good reason, which is hydrogen doesn’t last very long in the subsurface. The whole point of refinery is if we throw hydrogen, some long chain hydrocarbons, it absorbs the hydrogen, breaks them into smaller chain hydrocarbons. And for that reason, I think a lot of the hydrogen that has been generated from serpentinization or gets generated maybe from microbes or other things, actually will just sort of react in the subsurface and be hard to economically recover.
But it’s still cool to explore it. Going back full circle to the start, of course, some companies should give this a go, and fortunes will be made and value will be unlocked when analysts like me are proven wrong and you managed to do it. And so darts will be thrown at the dartboard, they won’t all hit. But still good to be throwing the darts.
Peter Tertzakian:
Yeah, it’s interesting. I mean, as a former geophysicist, explorationist, I think though you can see the skepticism on my face, given that you probably have to drill a couple of hundred wells that are quite expensive because the hydrogen would have to be deeper as well before you find one good one. And maybe I’m even being optimistic with that statistic.
Rob West:
One of the challenges is for me is that we know we can produce natural gas at $3 an MCF. We know we can, we do that quite a lot.
Peter Tertzakian:
We can do it even cheaper than that, yeah.
Rob West:
Well, but if we get hydrogen out at the ground at $3 an MCF because of the density of hydrogen, like eight times less dense than methane, $3 an MCF is already $1.25 a kilogram, but that’s already the cost of SMR hydrogen. And that’s a way to look at it where I guess…
Peter Tertzakian:
But you do have to factor in that the probability of striking a natural gas well is far more certain than striking a successful hydrogen well. In other words, you have to drill a lot more and a lot deeper to find hydrogen than you do natural gas. So I think it’s going to be a long time before it’s competitive. But, who knows? We live in a world of surprises.
Jackie Forrest:
Yeah, one of my summer readings was Pierre Berton’s the Klondike, and I’m reading about the gold rush and all the craziness in the late 1800s, the gold rush. I read your paper, and I’ve actually been seeing quite a bit more written about this in the last two or three months as well, it’s the dream of finding something like that is pretty exciting too. Right?
Peter Tertzakian:
No, it is. So Rob, before we let you go, you’re the statistics guy, is it more common to create cat videos or dog videos?
Rob West:
I mean, I am a dog person, but I think it has to be cat videos in the video frequency distribution. Yes.
Peter Tertzakian:
Rob West, CEO of Thunder Said Energy, thanks so much for joining us again, and I’m sure we’ll have you back. It’s been delightful to chat with you.
Rob West:
Back at you. Thank you.
Jackie Forrest:
Thank you, Rob. And thanks to our listeners. If you enjoyed this podcast, please rate us on the app that you listened to and tell someone else about us.
Speaker 3:
For more ideas and insights, visit Arcenergyinstitute.com.
