We’re back for round two of our deep dive with analyst Nat Bullard, with even more slides than last year.
This year’s annual slide deck from Nat has 200 pages on the key trends shaping decarbonization in 2024. Nat has worked as an analyst and writer in climate tech for two decades and was BloombergNEF’s chief content officer until 2022.
We’ve split the conversation into two parts. In this first part, Shayle and Nat cover topics like:
Catalyst is supported by Antenna Group. For 25 years, Antenna has partnered with leading clean-economy innovators to build their brands and accelerate business growth. If you’re a startup, investor, enterprise or innovation ecosystem that’s creating positive change, Antenna is ready to power your impact. Visit antennagroup.com to learn more.
Catalyst is brought to you by Atmos Financial. Atmos is revolutionizing finance by leveraging your deposits to exclusively fund decarbonization solutions, like residential solar and electrification. FDIC-insured with market-leading savings rates, cash-back checking, and zero fees. Get an account in minutes at joinatmos.com.
Shayle Kann: I'm Shayle Kann. This is Catalyst.
Nat Bullard: There's a persistent cheeky header that I keep using on these slides, which is markets respond to incentives. And then, in parentheses, again. People keep thinking that, when I say that, I mean that they respond to positive incentives. But, markets respond to negative incentives too.
Shayle Kann: Well, it's that time of year when I grab a cup of tea wishing it were coffee, sit back, put my feet up and read Nat Bullard's 200 page slide deck on trends in decarbonization.
I'm Shayle Kann. I invest in revolutionary climate technologies at Energy Impact Partners. Welcome. Well, we're back. This is Year Two of the now and future annual tradition where my buddy, Nat Bullard, and I dig through my favorite tidbits from his ridiculously long and data filled decarbonization trends deck. This one has it all. We talk power, emissions, finance, batteries, ESG, molecules, land use, other things. It's a two-parter. Here's Part One and we'll be back next week with the second half. Nat, welcome back.
Nat Bullard: Shayle, it's great to be back. Thanks for having me back again. I'm trying to count down how many times this is. I think this is number four.
Shayle Kann: Number four on the pod seems right. But, number two of this now annual tradition of me carefully cultivating and picking and choosing the best amongst your many, many slides on decarbonization that you put out each year. I counted. This is a 200 slide deck report. I don't know what you want to call it.
Nat Bullard: 200 even. Yeah, 200 even. Yep.
Shayle Kann: I picked 30, which is more than we're going to be able to get through over the course of two episodes right now. But, let's see how many we can get to. I also... I sort of categorized them, right? I put them into some buckets. Let's just do it. First bucket is what I'm calling things that are actually going quite well, coda, that you might not realize are going as well as they are. The first one actually is, to me, the one that I feel like I should have known but absolutely did not know, which is... and I'll ask you this. Before you had seen this data, if I had asked you what is the current trajectory of greenhouse gas emissions per capita globally, would you have known the answer to that?
Nat Bullard: You're asking probably the wrong person for that.
Shayle Kann: Because, you've been looking at this data.
Nat Bullard: Yes. I've been looking at it for a while. But, it's all part and parcel of what I had as sort of a subtitle on this earlier, which is like 'What's the deal with 1973?' Is that actually there's this profound year in not just the global economy, but in particular in the energy economy brought on by the oil shocks of the 1970s that introduced some really fundamental changes in the way that we approach deficiency and what we consume and things like that. The result is that, on a per capita basis, we had all greenhouse gas emissions peak in the 1970s. In particular, methane and nitrous oxide emissions were peaking in the 1970s. CO2 emissions didn't peak until well into this century per capita. But, all of them equivalent peaked back then. The reason I say I saw this in the 1970s as well in other data is that that's also when the energy intensity of GDP in the developed world peaked. They are related. The trick is that they haven't really gone down that much.
Shayle Kann: Right. No, it's an interesting line. They peaked in 1973. If you're looking at all greenhouse gasses lumped together, peaked in 1973. That's definitely the thing I wasn't aware of. They've been flat-ish, I guess slightly down basically over the past 50 years since then. But, that is crazy. Obviously, we've had population growth and that's why raw greenhouse gas emissions have grown. But, on a per capita basis, they've been more or less flat.
Nat Bullard: More or less flat. It's an important thing to note that it is possible to make changes, and it gets to one of my... One of my themes throughout this is that the only way to change these levels in the long run is to crank down on doing the good stuff, crank down on doing the things that make a big change. That's efficiency, that's things that have lower emissions, whatever. But, try to move these as much as possible. Because, now, you're in this sort of a race between population growth and inherent rates of decline per capita in XYZ consumption of whatever or emissions of whatever.
Shayle Kann: Right. All right. Let's talk about another thing that's going well. This one may be more that people, at least who are listening to this podcast, probably know to some degree. But, this is where you and I get to wax nostalgic for a minute. Let's talk about solar. In 2023, the estimate is that about 440 gigawatts of solar was installed globally. I think you and I should take a moment to contemplate that number. I was thinking, when I started paying a lot of attention to solar professionally in 2008 basically, I was looking at the chart and in 2008 we installed maybe five-ish gigawatts globally. Over the course of the time that I've been professionally thinking about solar, it has gone from five to 440. You predated me, right? You were probably back in the one gigawatt days.
Nat Bullard: I started doing this in 2007, so not predating you by much. I was covering the US starting 2008 at the same time you were when the projection for installations was like a hand loaded spreadsheet of discrete assets plus tallying up a few websites where people attract stuff like in the state of California. Yes, we basically moved from market... In the course of two decades, you've moved from installing in a day what you used to install in a year on a global basis in these markets.
Shayle Kann: The other thing that's remarkable about it is the growth in 2023 alone. We went from 252 gigawatts in 2022 to 440 in 2023. That is not a market that has hit maturity and is starting to reach its asymptote and flat line. That is actually continuing to increase its rate of growth.
Nat Bullard: It's one of the fastest rates of growth since early in this century. If you were to take just China's installations last year, which is north of 200 gigawatts in and of itself, it would've been the biggest year in any other year besides 2022. Yeah, this is a theme that we can return to later on as we can look into supply chains and things like that. But, not only is it looking like it's still beginning to go asymptotic, it's nowhere close to running out of manufacturing capacity headroom either, which is pretty extraordinary. And, it's a suspension of disbelief moment even for us crotchety old solar analysts to remember that no, no, no, no, no. It's quite possible to double this number again in the next couple of decades.
Shayle Kann: A single one terawatt year is totally plausible. In fact, we're currently in a dramatic oversupply situation. Module prices have crashed because we are currently super oversupplied despite the size of this market.
Nat Bullard: The suspension of disbelief for you, for me, even for other market practitioners is like remember that these things go in cycles, that these markets tend to oversupply themselves, which has a way of becoming a forcing function for the market to grow further. Yes, there's always I think a temptation to look for some kind of asymptotic limit here. But, every time that we've done that in a considered and rational way, the market is blown right through it and it's looked very silly in retrospect to impose a preconditional asymptotic limit to the way that this is going to grow.
Shayle Kann: All right. Let's crow about solar for one more minute and then we'll move on from it. Which is, later on in the deck, slide 66, you make the point that I think everybody knows practically, but it's still remarkable to see on paper. Which is that, over the course of the last decade, solar modules have both been consistently getting more efficient and cheaper, and both by leaps and bounds.
Nat Bullard: A decade ago or 2012, let's start, if 15% efficiency was your middle market module efficiency, it's now 21% efficiency. That's pretty extraordinary. I mean, that's a 40% relative increase in efficiency, which in and of itself would be yielding you a great deal more energy and changing the levelized cost of the energy you get from a system. But, at the same time, the module price has gone from a dollar and 9 cents to 12 cents. You've had this incredible move in prices as well and in efficiency moving up. Again, it's this testament to an incredibly fine grinding effort of improving on the margin in units of hundreds of millions. At this point, clearly in units of billions of modules every year.
Shayle Kann: One thing that I always point out though is that it is now true that actually looking at the cost of the solar module is less relevant than it has ever been historically in terms of the cost of the solar produced, specifically because modules have gotten so cheap. That cost down trajectory the modules have seen has not been matched by I think literally any other component of the cost stack of solar.
Nat Bullard: We come from the time... When we started off, the rubric was, well, modules are half the cost of the system. Obviously, that's not the case anymore. I guess you're having a 24 cent per watt fully installed utility scale system, which we don't. That's something that we've had to adjust expectation-wise. It's also interesting to note that many of the comments that I've gotten all in good faith is people saying, "That's not anywhere close to the price." I'm saying, I'm like, "Well, you are buying modules for a rooftop in California. You're not buying them for a utility field in Western China."
Yes, the realized price for many people was not going to be exactly at that level. But, nonetheless, those are the prints. That's what the market is seeing and it's important to recognize that, if module prices hit there and stay there for a while, that will be fine for the market's perspective. But, also, to your point, it's very worthwhile to note is that now the improvements have to shift elsewhere. You've got to get better at all of the other things that flow in. And, that are exposed to commodity prices too that you can't just sort of wish away.
Shayle Kann: Not just commodity prices but also cost of financing. We're going to talk about this more later because you have a slide on it. But, there is a rational way to reconcile the fact that, on one hand, solar module prices have continued to decline while, on the other hand, at least in the United States, the cost of solar PPAs has increased during that same period. That speaks to everything else. All right. But, before we get to bad things, let's talk about some more good things that are going surprisingly well, particularly lighting. There's two pieces to this and they're tied to each other. One is what type of lighting are we installing? The second is, what is the result in terms of the amount of energy we use for lighting globally? Let's start with the first one. What type of lighting are we installing now?
Nat Bullard: We are installing LEDs. More than 50% of all global residential lighting sales in 2022 were light emitting diodes, which is pretty extraordinary. Because, in 2010, it was 1%. We are of an age where there were technical and political debates about shifting to compact fluorescence from incandescent bulbs. What this suggests is, first, that the LED is obviated in a lot of that discussion. I think you can probably still buy CFLs somewhere. But, LEDs are a far superior option. And, we're reaching the point where they will probably saturate the market. I have a hard time seeing outside of very niche applications where you have any lighting that is not LED technology. But, that's a really, really steep leg up. And, it's also persistent. When an LED gets installed, it's going to be there. In my previous home, we installed about... I don't know. I think we installed about 50 LED, matched LED, bulbs in every can in the house when we moved in. And, five years later, I moved out having never changed a single one of them.
Shayle Kann: Yeah, I mean, two things that are remarkable about this to me. One, the pace. Right? When you scroll out here, and this applies to solar as well, that's 20 years and we went from 1% to 50%. The second piece which is related to that is that that's global. More than 50% globally is really impressive. That includes the global south. That just speaks to the pace of change that's possible, at least in some product categories. Obviously, lighting is a thing that's turning over more frequently than power plants, for example, or even vehicles. So, it's easier to see that quicker. Nonetheless, they're super impressive and the result of that, of course, is that LEDs are much more energy efficient. So, we have significantly reduced the share of global electricity consumption that is coming from lighting over that period.
Nat Bullard: No, absolutely. It's fascinating because we think about the LED is essentially a consumer discretionary type of decision with consumer durable timelines behind it. You make this decision by going to a big box retailer and looking through a set of light bulbs because you need to swap out bulbs. That's, these days, a decision that you make and pay for with a credit card. But, it has these implications as if you were buying an automobile in terms of the timeline that it's effective over. Yes. Now, lighting, and this is still totally fascinating to me, is not only a lower percentage of total electricity consumption than cooling, but even than low temperature heating in this century. It's come down from a share of more than a quarter of all the electricity demand at the start of the 20th century down to barely over 10% of our total electricity consumption from lighting.
Shayle Kann: That's a function both of the fact that lighting has become more efficient and also the fact that we have developed as an economy and are producing more stuff and using more energy for other things, obviously, or more electricity.
Nat Bullard: It's fair to say in the year 1900 there weren't that many things you could do with electricity to begin with.
Shayle Kann: How many data centers do you think there were in the year 1900?
Nat Bullard: Very few. I'm going to go with no, none rather. You just had a very limited use. Again, of course, you have to put in the fact that 10% of today's electricity consumption is slightly more as an absolute figure than 25% of the year 1900's electricity consumption. But, nonetheless, it's impressive. I like to highlight things like this because it shows you, A, where successes have been made and, B, where these new frontiers of work to be done have moved. That's definitely into cooling and into low temperature heat. These are areas that now need to be addressed with electricity, and the electricity within them need to be addressed as well.
Shayle Kann: All right. We're going to do two more in this category of things that are going pretty well and you might not have known it. The first one is on land use and deforestation. Which, I don't know, you hear a lot about the problems of deforestation. I'm not sure everybody recognizes that we've actually done a reasonably good job on a global basis of cutting that down. I didn't even mean that joke as I said it.
Nat Bullard: Fair enough. Cutting it down. Yes. Well, let's highlight within the data that I have here that I'm looking specifically at the Amazon. This is the place probably where the data are the richest and the time series the longest. But, yes, we had about 5,000 square kilometers of deforestation in the Amazon last year, which is like 60% less than it was in the year 2021 and much less, like maybe a fifth of what it was or even less, in the 1990s and in the early 2000s.
We had this unfortunate run-up in deforestation in the last seven, eight years in Brazil. But, that was to levels that was far lower than it had been in the 1990s. Again, the rates shrink because there's already a lot of deforestation that has already happened. The absolute number is cutting into an also absolutely smaller total area of forest. But, it's important to recognize the real number. Because, again, from a carbon cycle perspective, this number matters regardless of what it is and you want that to be as close to zero as possible.
Shayle Kann: Well, speaking of the carbon cycle, if you look at it from an emissions perspective, this is not just deforestation. But, one of the things, whenever you look at the emissions data and you're parsing out the different parts of the pie, there's always this forestry and land use bucket that's always very large but confusing to try to understand exactly what that means. But, if you look at the trajectory of emissions in that land use change category, they also have been heading pretty rapidly in the right direction it seems.
Nat Bullard: That's right. With some very big spikes and with the proviso from all of the researchers to do this, that data tends to be revised later on. But, yeah, we're down from eight gigatons of land use change emissions in the late 1950s and some spikes back up to more than seven gigatons in the late 1990s to about four right now. There's still a lot of emissions that's coming from this in and of itself that would be bigger than almost any country on its own. But, again, trending down and the success that we can look to. But, also, not trending towards zero. These are things that need to be moved further and faster and to be thought of I think also as a resource. This is another recurring theme throughout, is thinking of the land and how we use it and how we engage with it in a broader sense than just covering it with PV panels or growing corn on it or cutting down whatever's growing on it and burning it.
Shayle Kann: All right. Last in the category of things that are going pretty well is the trajectory of energy storage on the grid, which is another one sort of solar actually. I mean, it's less mature than solar obviously, so it's like earlier in that adoption cycle. But, similar to solar, the curve is bending upward at the moment on a global basis. It is not flattening out. And, in 2023 in particular, it seems that basically nobody predicted how much energy storage was going to end up being installed on the grid on a global basis.
Nat Bullard: Yeah, my former colleagues at BNEF gapped their projection up by almost 30 or 40% I think if I look at the numbers, at a really substantial leg up on what had been anticipated, to about a hundred gigawatt hours of total storage in 2023, up from like 30 or so in 2022. Again, it fungos a very solar logic to it as well that I think you and I would find familiar. You don't want to pattern match too much. But, a highly distributed resource with manufacturing economics on one end and project developer economics on the other. Where there's clear market signals people can deploy fairly fast and move assets into the grid fairly quickly. And, to an extent changing prices also creates its own market.
Yeah, it's a big deal. I mean, again, we should be thinking forward to this on when do we get to our first terawatt hour year at this point. That's where we will start to see really meaningful contributions. We'll also start to see changes I think once we reach that scale in terms of what people are doing with it. It won't just be four hours of storage. It won't just be for frequency regulation of voltage support, stuff like that.
Shayle Kann: Right. Okay. That's our category of optimism for the day. Now, we're getting it into some stuff that's less obviously positive but interesting nonetheless. Category two, I just called money, money, money. All right. We're going to get a little into the weeds on this one. But, I haven't really talked about it on this podcast before and it's important, which is what's happening with transferability of tax equity in the United States. If you already know what I'm talking about, then you don't need this explanation. If you don't, then this isn't going to be sufficient for you. But, basically, the Inflation Reduction Act did something important, which is that for the first time you can transfer your tax equity credits if you are generating them via a renewable power installation or an energy surge installation.
It used to be that you had to somehow consume those credits yourself or create some complicated financing mechanism for your project so that your tax equity investor was an owner in the asset directly. Now, they can be transferred and that's a new thing as of the last year. We're getting some interesting early data on what's happening with those transfers and what they cost. Because, somebody else has to buy those credits and they're not going to buy them for a hundred cents on the dollar. The question from day one was how are buyers going to be thinking about that. That matters because it impacts the economics of the project at the end of the day. With one year of data behind us, what do we know about the cost of transferring tax credits?
Nat Bullard: What we know so far, and remember it's a thin field, but it's in a very heartening way, growing very rapidly. And, also, becoming increasingly data rich. We're seeing what I think you would hope for which is that the smallest ticket sizes, so the smaller the transaction, the deeper the discount to par value is, which is a normal kind of thing for having to deal with the aggregation challenges and the risks involved in smaller assets. The bigger the purchase price is, the ticket size, the smaller the discount to par.
For projects that are, or a transfer of under a million dollars, you're getting like an 84 to 86 cent price as opposed to what you say it suggests, it's a hundred cents on the dollar, which gives the investor a quite high IRR. Like a 16 to 19%, sorry, ROI rather, return on investment. When you get all the way up to say a $50 million transfer where you're going to be competing with traditional tax equity, but not only with traditional tax equity but with other risk-free investments like say treasury bills, you're in the 94 to 96 cent price point and a return on investment of four to 6%.
This is data that basis climate published late last year. Crux has also published some similar data. It's very useful to see all of these numbers because this is how... This is, A, what you want to see in an evolving market and, B, I think the sort of pattern that you would expect to see. It would be unusual if you didn't have these spreads between par value, spread of par value between little and big projects and if you didn't have changes in return on investment accordingly.
What I like about this is that it gives a spectrum of risk appetite and return appetite for investors. If you're a retail... not retail, or you're a small specialized investor, you may be interested in doing projects under a million dollar ticket size because you're willing to get a 16% return on investment. If you're a very large investment bank, you're probably not expecting to deploy 50 to a hundred million dollars and receive something that's three times the risk-free rate. You're therefore looking at pricing points that are probably closer to where T-bills are printing.
I think this is great. A, more transactions, the better. B, the more transparency, the better. Because, this is a market that looks like somebody will be able to get behind. I think it would be really healthy to have a much broader market of participants than the specialists who can do... than only rather the specialists who've been doing it since you and I started doing this work. It is fascinating, no judgment statement. But, a lot of the players in tax equity today, not just the institutions but the people are the same ones that you and I would've met for a cup of coffee at a conference in San Diego in 2009. It's really valuable to see this expanding and taking on a platform basis as opposed to a very bespoke basis. The analogy would be that it's basically moving from the bond market to the equity market. Or, it's moving from over the counter trades to exchange trades. It's not quite like that yet, but it's getting towards that level of transparency and liquidity, hopefully.
Shayle Kann: Speaking of cost of capital, I liked in slide 50 it's not like new market data, but it actually makes a really important point that is salient to what's been happening in the market the past couple of years. Which is just laying out, at different costs of capital, how much of the total cost of any given project is made up of your financing versus CapEx and OpEx together. As you point out, and this is just taking for a renewable project. We know the thing about renewable projects is that they're all upfront CapEx basically and the OpEx is very small and they're particularly sensitive to those costs of capital. As you point out, with a 4% cost of capital, 75% of the overall project cost is born out of the project itself, CapEx and OpEx. 25% is financing. But, if you're up in the 12% cost of capital range, now 60% of your total all in project cost is your financing. I think that goes a long way toward explaining why it's been such a big deal that interest rates went way up over the past couple of years.
Nat Bullard: Absolutely. We had this sort of holiday from rates history for almost 10 years in which the cost of capital at the risk- free rate was basically zero and you could borrow long for long-dated assets at extremely low rates. And, you could not have to think too much about the impacts of the cost of capital on your asset because the cost of capital was so low. Now, unwinding that, getting back towards an actual rates expectation of three to four to five, 6%, and that's in developed markets, means that, yes, the cost of money is a huge lever on the cost of your delivered energy from a project. And, it's very tricky because there's only so much engineering you can do to get yourself to the lowest rates you want.
When the rate is zero, you maybe don't even need to engineer that much at all. You get as close as you can to it and you're in a realized rate of a couple of percent. But, it's going to be unlikely to go below the risk-free rates of capital. You're always going to have some sort of buffers on top of them and some of them could be very substantial as you see in some markets. The adders onto the risk-free rate could be adjustments for currency risk, adjustments for sovereign risk.
You'll be at the point where your cost of capital is 11% or 13%. It does happen in markets like South Africa or India. The single biggest never you could have to move that is the risk-free rate itself. But, the real challenge for the industry is that it's hard to think of a more exogenous force than that. There's not really a way that, as a developer, I can go in and be like, "Yep, thanks to my good work, I've managed to move the secured overnight financing rate down by 75 basis points." You are almost purely a taker on it and you are subject to and captive to whatever's happening in the rest of the economy.
Shayle Kann: Right. The last thing in my money, money, money category that I thought was actually really interesting analysis is in slide 55 you're talking about what it would've looked like had you done a long-short strategy, which you should explain what that means for anybody who doesn't already know, where you are long clean energy, short traditional energy, and then how the returns on that strategy would've changed over the past six years or so. First, just briefly explain a long short strategy. I'm sure most people know. But, just in case. And then, how would the returns have changed?
Nat Bullard: Long-short strategy is that you are looking for growth and you're positioned for growth in one basket of securities, and you are expecting a decline in another basket of securities. In this case, the long-short, and this is research from Bernstein, says the iShares global clean energy ETF, you go long that and you go short the iShares global energy ETF. And, look, for three years, pretty good outperform. 2018, 2019, 2020, you had 17, 13 and 160% outperform excess return from this long clean, short dirty strategy. Which kind of unwound. It would've really stung in 2021 when that strategy went to a negative 77% excess return. That's some sophistic language to say you lost quite a bit. Negative 29%, negative 40% through September of 2023.
Look, I mean, there's a lot of stuff going on in here. One thing valuations came back down to earth for a lot of these clean equities. There was a lot of cleanup happening within the global energy ETF components. So, companies de-levering which definitely, in the oil and gas sector, is going to mean that they're much better able to return cash to shareholders. You also had a lot of buybacks. You had companies that are propping up their share price by devoting cash towards removing their total number of shares and therefore driving prices up. It's complex.
I think... I didn't do this math here. You could take this math starting with a rebased value of a hundred and see if you made money over the long run through this strategy. You might have. But, it would've been a bit of a roller coaster to get there and probably not the way that you would want to get towards your S&P 500 index equivalent return. You probably don't want, for the sake of being an indexed investor, to reach your 10 to 12% annual return or whatever that number might be by being up 160% one year and down 77% the next year.
Shayle Kann: There's also an interesting philosophy embedded in that strategy, right? That strategy effectively says I expect clean energy to win at the expense of traditional energy. Now, in the long term, that probably is a reasonable thing to bet on. Ultimately, if you expect a 30-year positive trajectory for clean energy, of course that's going to come at the expense of traditional energy. In the short term, it's not entirely clear to me that that's true, at least not in an equities context.
There was a period of time when actually both of them were... The energy sector was performing really well and you could have defined energy sector as traditional or clean and either way you would've been looking good. That turned a little bit when interest rates rose and oil and gas prices rose and all that. That then made the traditional energy perform much better than clean energy for the past couple of years. But, my guess is, not having done this analysis, you probably would've done better if you had just gone long clean energy and then forgotten the short position on traditional energy.
Nat Bullard: That's probably... Well, I don't know. I couldn't speak to that. But, it would be simpler, that's for sure. I would agree completely that there is this false sense of correlation between if long this than short that. I agree with that very much. They don't necessarily correlate that way. In analog to this is the assumption say, six, seven years ago that you will divest from fossil fuels and therefore put all of your money into clean energy equities.
In fact, it was 10 years ago now that I wrote a white paper about all of that and tried to emphasize, "Hey, if I'm moving $5 trillion out of oil and gas and coal equities at the time, I don't have $5 trillion worth of clean energy stocks to put it into." And, I also have particular types of return thresholds that I'm used to. I might want a lot of yield, so I'm probably going to go to real estate. Or, I might want a lot of liquidity, so I'm going to go into tech. Or, I want a lot of diversification, so I'm going to go into any number of other indexes. The one does not follow from the other. That's a larger story slash lesson about divestment that we could discuss at another time. But, yeah, going long one thing does not necessarily mean being short something else.
Shayle Kann: All right. That wraps up our category two, money, money, money. Let's move on to category number three, which is trends in technology. Starting with we talked about how big the battery market is now. Let's talk about who's making those batteries. Slide 63 makes the point that I think probably folks appreciate generally, but it's still fairly remarkable that it is true, which is that all 10 of the top 10 largest battery cell manufacturers are now in Asia.
Nat Bullard: A goose egg for the rest of the world in the top 10. In fact, all of the others are 6% in total.
Shayle Kann: You mean the total aggregate Asia-based battery cell manufacturing is 94 or close to 94% of global production?
Nat Bullard: I believe so. There's just not a lot of stuff happening elsewhere yet. There's a lot under construction. There will be some changes there. Remember, this market is very, very big. You can still have really, really big scale happening in the 50 to a hundred gigawatt hour, probably capacity capabilities in other countries that get stood up. But, yeah, it's China in first and second. It's South Korea in third and fifth. It's Japan in fourth. And then, it's China, Korea, China, China, China from there on as you go down the list. I believe it was also the first year of producing a terawatt hour worth of batteries, which is notable. But, yeah, this is not a new story. I guess what's notable is that it hasn't particularly changed that much. I could go back and compare this to last year. But, this is directionally pretty much exactly where it was in 2022.
Shayle Kann: One thing that is different about batteries thus far from solar, solar also moved all to Asia. But, it ended up being in solar almost exclusively Chinese companies. Now, over time, those companies maybe were producing in southeast Asia because of tariffs and all sorts of other things. But, the epicenter of solar manufacturing moved entirely to China. Now, in batteries as of today, South Korea between LG, SK, Samsung, and then Japan with Panasonic, they still have a pretty prominent role. It'd be interesting to see whether that sustains over time. Because, obviously, the big growth companies, the big winners over the past few years have been CATL and BYD, who are the numbers one and two now.
Nat Bullard: That's right. To see where are those, what's the go-to-market outside of Asia as companies expand, as they build new capacity in other places, who are they going to build with? Who are they going to partner with? Some of them might be Northvolt's file and be building their own stack. But, others will be partnering with companies and who are they going to partner with is going to be the real question.
Shayle Kann: The other thing to me that I think maybe not everybody appreciates is that actually technology changes pretty quickly in battery world, at least it has as of late. The perfect example of that has been the pace of the shift to LFP batteries for electric vehicles. You point out in the next slide, in 64, LFP went from 5% of EV batteries in 2019, not that long ago, to 42% in 2023. That's a very quick turn for an industry that requires gigascale manufacturing to make a meaningful share of the market.
Nat Bullard: This is a very big deal. This is really important to see that lithium iron phosphate batteries, yeah, went from de minimis share of market to almost half in five years of time. This is largely a function of the Chinese EV market and of deployment within that particular market. But, I have to say that it's unlikely that this capability and this chemistry is going to be ring-fenced to China forever. The spill-on effects of having LFP batteries elsewhere in the world in other applications is going to be a really big deal. I like this quite a bit because it's something where I really have to interrogate my priors from solar, which is one chemistry tends to sort of wipe the floor with another and that's it. Or, one configuration tends to blot out everything else with the exception of some really application specific things.
In batteries, I think we will see a world where you've got some separation based on use cases. There will be the NMC batteries that get used for hypercars and for aviation, and maybe the LFP battery becomes much more common for your bulge bracket or middle market automotive applications. And then, you'll have batteries that use sodium, for instance, that will find their way into the system as well. It's another case wherein the exceedingly fine grind of volume manufacturing has a way of moving capability pretty relentlessly.
The really great thing is that, when these things leave the lab bench and enter into volume production with manufacturing, is you get visibility on them. I think it's really important as a sort of aside to watch the market data as opposed to what's coming out of just pure research. The research is fascinating. It's like an advanced look on the future. But, the data wherein a large company makes a commitment to do something at scale means that they think it's going to work and it is likely to at least find some application where the market contested.
Shayle Kann: Okay, speaking of markets moving quickly but for different reasons, let's jump to Slide 71, which is about residential solar in California. Basically, I think anybody who's in the residential solar market knows that this has been going on. But, California revised net metering policies basically in a way such that when it has eviscerated the market, the overall volumes of residential solar have gotten killed. But, to the extent that residential solar is still getting installed, all of a sudden it is largely used in batteries.
Nat Bullard: That's right. There's a persistent cheeky header that I keep using on these slides, which is markets respond to incentives. And then, in parentheses, again. People keep thinking that when I say that, that I mean that they respond to positive incentives. But, markets respond to negative incentives too. Yes, basically, if you no longer have the policy gifted economics of net metering working in your favor and you still want to do solar, well, magically, you now need to attach a battery to your asset in order for it to make any economic sense.
Again, this is a case where I feel like this is a very intended consequence. If I'm a policymaker, you should know very well that this is what was going to happen. If you introduce this policy, you're going to significantly impair a market. But, the flip side of that is, for the admittedly smaller market that persists in residential solar in California, it now looks very different because attaching batteries makes economic sense. You'll start to see new optimizations being built around that. But, yeah, it's very, very fast. It was like a couple of percent of systems had a battery attached in the middle of 2023 and now almost 25% do.
Shayle Kann: For some installers, it's well over 50% now.
Nat Bullard: I'm sure.
Shayle Kann: It's just going to be the norm in the market just because economically it makes way more sense now. It's the only way that makes sense.
Nat Bullard: Magically, it makes sense. We think about markets being willed into being by a new subsidy coming into place or a new subordinate mechanism. But, they're also willed into being by other mechanisms going away.
Shayle Kann: Staying on the battery theme for a second, we're going to jump to talking about the cost of batteries and what has changed there. This has been another interesting market dynamic that we haven't talked about a whole lot here. Jumping all the way to Slide 145, lithium prices have gone on quite a wild ride over the past couple of years, and then have had a direct result on battery prices over that same time.
Nat Bullard: Absolutely. Yeah, the prices went up by a factor of about 10 over the course of two and a half years from January, 2020, to the middle of 2022, late 2022, and then collapsed. Again, this is a sort of supply demand thing. There are elements within the market of how products are traded and things like that that are a little bit outside my gift to explain. But, the effect for battery makers is that you had to absorb or blunt this massive run-up in the cost of lithium that then collapsed and will eventually see its way through to the product that you deliver.
Contracts, depending on the sort of elasticity or stickiness of contracts, you may find yourself in a position where you have pre-sold things based on prior prices that you can now use to your advantage with the new prices of inputs. But, in the long run, it will have to mean that battery prices probably come down and get closer towards their cost of goods sold. It'll be hard to drive a persistent wedge in a competitive market like that forever. Yes, good for the battery market. In a complex signal, I would say for the necessary exploration that might need to come to have more lithium resources come online. It'll be an interesting signal for somebody who might be preparing to go plan new mining operations. But, again, part of a complex interplay then back again with, say, batteries that use NAS lithium or use no lithium, to think about how that's going to play in the long run.
Shayle Kann: Yeah, that means the classic the solution to high prices is high prices, the solution to low prices is low prices. The question is the timing of both of those things and how long it takes for all the incentives to flow through. Speaking of batteries, another thing we haven't talked about here that you have the data that I've been wanting to point out for a while, and indeed have had to point out many times because we're on the investment side, we're looking at lots of companies that want to be doing battery recycling. There's a fundamental challenge with wanting to be a new battery recycling company right now and that has to do with supply and demand of recyclable batteries. Walk us through where we are on the supply demand equation for recyclable batteries.
Nat Bullard: This is Slide 150 and it's some just fantastic data from the folks at Circular Energy Storage. Like you, I had been looking for this for a really long time. The simple fact is that we have vastly more material recovery capability than we do supply of things to be recycled. The ability to recycle is way outstripping the material that is coming in to be recycled and that is likely to be recycled pretty much all the way through 2030. There's a lot of nuance in the way that this might evolve in the market. First is you could change the supply side by having end of life battery supply go way up. The end of life of batteries could be happening faster than people expected or in greater volume and that could push up closer to what the ability to recover material is, getting supply and processing capability closer in line.
But, the other one is, what is the end of life for the battery? I think to me this is the biggest question. The unwritten question within this chart is, the end of life for somebody's high performance Tesla battery, driving it in Orange County, California, is a different concept of end of life than it might be if it's being shipped to Eastern Europe or Southeast Asia where people are going to continue to use it and try them. We had this question decades ago with Prius batteries. Do they get chopped up and turn into scrap or recycled, or do they go on to be somebody's grid application, or off-grid application rather? A really, really big question.
Shayle Kann: Yeah. But, the numbers are stark here, is the point that I would make. According to these numbers, and this is looking at announced capacity for battery recycling for end of life processing and material recovery versus projected end of life batteries in production scrap. Either of those numbers can move, as you said. But, according to the numbers that Circular Energy Storage compiled, in 2030, will be at about two million tons per year of combined end of life batteries in production scrap versus 10 million tons or so of capacity for material recovery and processing.
There's a five X gap there, which is like... That's a problem. That means on average those material processing and recovery facilities will be operating at 20% capacity. Now, in reality, a bunch of them are not going to happen because of this and some others may get to a hundred percent. But, to me, if you are thinking about doing battery recycling right now and you're not already at scale sourcing feedstock, that is question number one, two, and three. Because, you are headed into a market that is pretty clearly going to be oversupplied from your perspective.
Nat Bullard: A hundred percent. It's always worth... To reiterate, it's worth remembering that you're competing in that end of life category. Not just with every other end of life battery that might be recycled, but with other uses with end of life question mark. Like, what is the end of life for that product? From a material efficiency perspective, it's still going to be way more efficient to keep using that battery as long as you can than to send it off to the battery boneyard.
Shayle Kann: Right. Okay. I think we've gotten through roughly half of the many slides that I preselected here. Let's call it for today and we'll come back next week and pick up with my next set of carefully curated slides from your deck.
Nat Bullard: Sounds good to me, Shayle.
Shayle Kann: Nat Bullard is a longtime climate tech analyst and writer. He is formerly of Bloomberg now doing his own thing, including writing this deck. This show is a production of Latitude Media. You can head over to latitudemedia.com for links to today's topics. Latitude is supported by Prelude Ventures. Prelude backs visionaries, accelerating climate innovation that will reshape the global economy for the betterment of people and planet. Learn more at preludeventures.com. This episode was produced by Daniel Waldorf, mixing by Roy Campanella and Sean Marquand, theme song by Sean Marquand. I'm Shayle Kann, and this is Catalyst.