Go back

95 | What Does CDR Actually Cost in Europe? - with Hansjörg Lerchenmüller and Eadbhard Pernot

from The Carbon Removal Scoop

37m 55s

95 | What Does CDR Actually Cost in Europe? - with Hansjörg Lerchenmüller and Eadbhard Pernot

The EU Commission’s carbon removal cost estimates in its ETS review rely heavily on a limited set of sources—McKinsey, NEGAM, CDI for AI, and Ramble—many of which use outdated, global, or speculative data. For bio-CCS, costs are modeled at unrealistically low levels due to assumptions about energy revenue and transport, ignoring real-world biomass sourcing and infrastructure costs. Project developers in Europe observe actual costs ranging from 150 to 340 euros per ton, with significant variation based on project scale, location, and feedstock. Direct air capture (DAC) costs are similarly uncertain, relying on non-EU case studies and flat learning curves without justification. Biochar costs are similarly skewed downward, with global data ignored in favor of European-specific realities. Experts emphasize that current models fail to reflect real project economics, supply constraints, and infrastructure risks. A critical gap is the lack of transparency and real-world data. To improve credibility, the Commission should adopt transparent, up-to-date data from CDI for AI and encourage all stakeholders—including buyers, suppliers, and developers—to contribute real project cost data. This would enable more accurate, realistic, and actionable carbon removal cost modeling for EU policy.

Transcription

6721 Words, 36910 Characters

English
Hi and welcome to the CDR Policy School, where we unpack carbon removal policy in 30 minutes or less. My name is Eva Tamman. And I am Sebastian Menhoff. We recorded this conversation on Monday the 17th of August. It has now been a month since the European Commission published its EU Emissions Trading System review proposal and the accompanying impact assessment. In this episode, we explore what other commissions cost estimates for biocCS decks and biochar are holding up against what is happening in the market right now. Sebastian has recently published a deep dive into where the numbers are coming from and we have two guests with us with a first-hand knowledge in the real project economics to comment on it. Everett Pernault, the Executive Director of Carbon Management Europe, focusing on biocCS and decks and Hans Hyrk-Lersen Willer, the Founder and Chair of Biochar Europe, focusing on biochar. With Everett and Hans Hyrk, we covered a lot of ground. We started off by discussing how these costs actually feed into the broader proposal that we are discussing here and the role that they actually play. Then we dive into the specific assumptions behind the biocCS decks and biochar carbon removal costs that we find in the impact assessment, talking about transport and storage, feed stock, energy costs, everything that is relevant. And finally, we also covered the potential of these different technologies in the European carbon removal portfolio. Enjoy the show. Hello, Everett and Hans Hyrk. Thanks for joining us today. Thanks for having us on. Great to be here. Yeah, hi, both. Super excited for this session. Yeah, I was telling Everett just as we, as I joined the studio, how excited I am for this session. Yeah, I've been nerding out on the cost modeling and forecasting in the ETS proposal and I'm so thrilled today to talk to two industry experts who can really help stress test some of the assumptions. And maybe to kind of set the scene a little bit, why are we even doing this obviously the ETS proposal came out recently and in the impact assessment, there was a pretty big section on the costs for CDR, especially for biocCS and DAX, but also for biochar carbon removal that the Commission forecasts. And that is a big impact because they broadly assume that the cost of an emission allowance in the 2030s will roughly overall be equal to the cost of carbon removal. And if that were not to be true, things can come out of balance pretty quickly with implications that that would have. That's where we're discussing it today. And what I wanted to start with is kind of the, from my analysis, we're going to go technology by technology by CCS, DAX and then biochar debate each one individually. But I first wanted to give some overarching framing of how this modeling has actually been done. First, if you really dig deep, all the numbers essentially come from four sources. It's a McKinsey 2023 report. It's a ramble ecologic study from 2025. It's the NEGAM study from 2023 and CDI for AI data from 2023. So all the data we see comes from these four sources, which actually makes it quite easy to talk about the underlying assumptions. - Wait, CDR for AI from 2023? - Yes, because it comes from a study that used the data in 2023. But I have 2024 to 26 data with me today. But in the impact assessment, it used 2023 data, the newest. So first, how does this overall work? The commission looked at the costs of CDR between 2030 and 2040. For DAX and bios CCS, there's a yearly breakdown for each year. For biojure carbon removal, they only looked at 2030, 35, and 40. For each of them, there's a low, medium, and high cost scenario. The low and high is taken from other sources, and the medium is always just an average of the two. Which is why it's so important to look at the low and high numbers because they have an outsized impact of the medium expected scenario. A couple more things to note is that inflation adjustment was not consistent. That's not a huge point, but I do think it can have an impact of plus minus 10%. We probably won't talk about that today, but I thought I'd flag that costs and prices get mixed up quite a lot. Which again, has some impact, but I don't know how deep we will go into that today. And there's very little explanation for how some of the EU commissions own numbers came about, because a lot of the studies I mentioned don't have, actually none of them have, 2040 numbers. So the 2040 number is always kind of an extrapolation by the commission. And they've often just continued kind of the assumptions that happened until 2035 or 2050 and inserted 2040. I thought I'd flag that at the beginning, because that's a commonality across all these different approaches. But yeah, before saying anything, I do want to say to cut the commission a bit of slack, the data out there kind of sucks in many ways. And so this is something that that's why we're recording also this episode today is there's not a lot of great data out there. And so the commission kind of worked with what they had, and I hope that we can we can help with some of the discussions we're going to have today. Before diving into buy CCS, maybe overall when any of you when you were looking through this impact assessment, and you saw those numbers, what were your first kind of got reactions or impressions? I think like there's a great quote that George Box has, which is that all models are wrong, but some are useful. So I think actually I'd like to coin my own one, which is that all impact assessments are wrong, but that's not the point. The point is really when you're trying to produce an impact assessment, you're trying to assess kind of what it is that's the best based on the best available information that you have, what it is you're trying to achieve. And I think with the ETS proposal as it is and everything that really comes in the legislative text, which is ultimately what really matters, I think, and I think we can all agree on that. Great, I mean, for carbon removal before 17 to July, which actually was my birthday, so you could say it was a great birthday present, the legislative proposal, but you know, on the 16th of July, if you ask people, what is there to support carbon removal at the EU level in terms of funding, in terms of offtake and sort of at least in the future, there was nothing. And yet on the 17th, we have 250 million tons, and I think we as a community, especially, you know, those who are advocating for carbon removal, I would hope can all agree that we should applaud the commission and the efforts that they've taken to kind of promote permanent removals and using compliance markets or least funding coming from compliance markets to get there. But so I think that really like we shouldn't get ahead of ourselves and just sort of criticizing the data that we know is wrong, but I hope over this episode, we can kind of go into details of like what we should do in the future, maybe based on that uncertainty that we have. Definitely. And that was also basically my, my first reaction was it's great it's there that the numbers will need revision. So I am much more at home in industrial reality than in policy talks, so much less reading policy papers that I'm talking to industrial people. And what I observed over the last years is that they are kind of three sources of data on prices, which lead to sometimes unrealistic results. The first source is scientists, which is very often and I have to say that I'm I do have a bad scientific background work a lot with science, but there is kind of a culture of reading papers and writing papers and then quoting papers in your own papers. The second source is the startups that just promised Manhattan to their investors and that obviously leads to very low prices that seem to be feasible. And the third source of low prices is the oil and gas industry, which has an inherent interest to say, look, it's not that bad. We don't have to limit ourselves in emissions, we can solve it in 10, 15 years and it's even cheap so we don't have to do that. And then in 15 years, I promise the result will be whoops, it wasn't as cheap as we thought, but still we had emitted much more that we should be that's that's my concern on the numbers. This being said, you have to start somewhere, it's great. We have the program, but I think we have to be very careful on being realistic on what prices can be achieved and prices not cost. Are there anything from you when you look through the numbers? No, I'm just listening to Hans Ergen thinking, where did all the gas say that bio CCS and tax will be cheap, but I haven't seen that happen in front of my own eyes, but maybe let's jump in and we'll find out. Let's jump in. On bio CCS, this is by far the most important when we're talking about this proposal, because bio CCS constitutes 90, 95, maybe more percent of what is actually being proposed here realistically. And the other thing that was interesting with bio CCS was that the immediate reaction when speaking to project developers was like these numbers could be off by a factor of two. So not just a little bit, but like really off also the commission expects bio CCS costs and emission allowances to kind of cross paths in 2036 to 2038. So the bio CCS would actually become a revenue generator for the program because it will be cheaper than the emission allowances. So it's really important to dig into detail here. So let's look at the specific number. numbers. There are three numbers that have a huge impact here. First, the low scenario starting point, which is 172 euros in 2020. This is taken directly from the NEGAM study. And it's very interesting when you look at it because it's actually 337 euros, but they assume 165 euros of energy revenues. This is how you land at 172. So the key takeaway is almost half of this cost is essentially offset by energy revenue. So there's a huge component of essentially higher energy prices to make this happen. Interestingly, they also assume only 38 euros for transport and storage. I'd love to hear you take on that. And 105 euros for feedstock cost per ton. What was interesting is that they essentially just modeled this out and then they calibrated it against the statement made by the company Drax that we cannot verify because it's not linked, it's not public, which was apparently made at a conference in 2022 and went in a similar direction. So for me, this was pretty wild to essentially see they put together some numbers. They heard that a company in 2022 said a similar number at a conference. And that's how they said, this is probably how much it costs today. So first number 172, which is the low cost scenario starting point. The high cost scenario starting point is 314 euros in 2024, again, then modeled to today. And this just comes from CDI for AI. This is essentially transaction data that happened between 2019 and 2023 on CDI for AI. So it's pretty old data. It's global data and it includes a bunch of subsidies. Now you have these two starting points. So the low one is pretty low. The high one is also pretty low, which squeezes the whole medium scenario very low. And then this is what I found fascinating in the biocCS one. They take an endpoint, a 2050 endpoint of 153 euros and all three scenarios track towards the same endpoint, whereas for other technologies, they kept parallel tracks. And this endpoint comes from a consultation with nine experts where they were asked how much it costs in 2050. And it's basically they're all named it's seven academics plus one representative from Draks and one from Stockholm XDG. So this is how these three numbers come together. Maybe also worth pointing out that BECs and biocCS is equated across the entire analysis. And so yeah, there you have it. A lot of assumptions on energy costs on feedstirk, on transport and storage and also on global prices. Let's dive into it. I'm going to dive in with something completely different. I'm going to talk about your shirt. I have a question for you. So it looks like a really nice shirt. I don't know how much it costs, but you probably know the price. But how much does a shirt cost? I don't know how much I spent for this one, but say 50 euros. Okay. That's a decent size shirt. Let's say, I don't know how much shirt costs because quite honestly it all varies, right? Like you have got my shirts that have made of great material like silk or something. Then you've got like really, you know, cheaper shirts that maybe have lower quality and can be less expensive. So when you ask a question like how much does biocCS cost or how much does BECs cost? The answer honestly is like, I don't know. And I don't think anybody does. But that's again, that's not really the point. The point is when we're looking at these different methods and even within biocCS and I suppose one of the issues that we certainly identified when it comes to these sort of cost assumptions around even just biocCS as a family of technologies is that they're inherently different. So you have like post combustion power BECs, let's say, let's say you're doing capture from the flue gas of a co-fart heat and power plant. Let's say, you know, stock on exeges as an example or the earth that are doing in Denmark. That's like chalk and cheese when you're talking about something that's separating CO2 from biomethane upgrading or from ethanol production at a biorefinery. You know, not only are you talking about different flue gas composition, but you know, in one something is being burnt and in the other it's not. And this also impacts then when you're thinking about the infrastructure, which is, you know, another point that you raised 38 euros a ton for transport and storage costs. Again, you know, it's a very, very low number, but, you know, I can think of cases in which that could happen, at least in Europe. You know, I think about Danube carbon storage and Hungary. They've got a biorefinery of nearly half a million tons a year separating CO2 that's basically pure and they're looking to inject into a saline aquifer that's onshore about, you know, a kilometer or two away. Their headline costs at least what they say publicly and everywhere that they talk is, you know, they will happily do this for less than 50 euros a ton all in. So their TNS costs are quite low in this regard. But if we're looking at offshore storage in the North Sea and we're, you know, trying to build a very large pipeline because the only way we can get there is through scaling then then yeah, maybe the upfront costs for that will be quite expensive. But again, it depends on where, where your site is, how big it is, what you're doing with it. And I think one of the other points I want to make as well about about buy a CCS and costs is there's a few different paradoxes that are there. So, for example, one of them is not only if you're talking about sort of existing plants, but let's say if we're talking about, you know, adding on a capture plant, especially at one that's doing post-combustion, you probably will have to burn more biomass to power the capture plant and to do all of that process. And of course, if you're building a new one, then you're going to have to source that biomass. And in many cases, the sourcing of the biomass itself is a very, very large factor in those costs. You know, at some point, it could even be up to half of the total costs of the entire facility, including the capture plant. You know, and that might be for your project, which could be, you know, deploying in the in the early 2030s. But as we go on, future, you know, okay, you get the learning rates that can help to bring costs down, right? Because we've got technological improvements, we've got, you know, learning rates and savings coming through deployment. But there's also a run on biomass. That means that there's much more constraints. And so the price of the marginal buyers is suddenly higher, which means that then your cost inputs are also going to go up. And at the same time, when we're talking about, you know, energy being sold. So for example, you know, then I'm left with a situation where the energy or the electricity in this example or the heat that I could be selling, I now have to power my capture plant to produce CDRs. So a higher energy price also means a higher opportunity cost if I'm deciding to do CDR. And so all these decisions, you know, the decision, let's say, to invest or to build a capture project and our to do CDR in the first place, it's basically the the agglomeration of various different decisions that all came to that point. And for some projects, it's a lot easier than others, especially for the smaller ones where, you know, it's essentially you're not requiring additional purification, you're not acquiring any changes to what you're doing already. And actually, there's a lot of low hanging fruit that's out there. Can I interrupt you there because there's quite a few things we can already disaggregate, and then sure can come back to it. But for example, you mentioned the biomass, right? And the biomass cost and that it could take up half of the overall cost can actually jump to Hansiuk because I know you know quite a bit about biomass as well. What's your take on how biomass in Europe and the cost of it, especially, might evolve over time? Yes, that's the $100 billion question, the cost of biomass. And it will be driven by how the energy markets in Europe evolve. So we are in the middle of an energy revolution with wind and solar taking over the base load power together with batteries and load management. And that will be huge. So we we will see a tremendous change in the energy markets. So what happens with biomass instead? My personal view is that biomass should not be burned at the first place. I wrote once a white paper that's quite a long time ago and there I realized and made the statement that any combustion of biomass is a missed chance to create a carbon sink. It could be a carbon sink because you're using the wood in buildings or it could be making creating a carbon sinks with biochar. So it's very, very difficult to project how biochar biomass prices are going. When I do assessments, I assume it'll stay roughly the same because there will be new consumptions of biomass for bio-economics like materials, but there will also be biomass going out of combustion definitely. And out of in Europe, for example, the paper industry, they will free up biomass. Can I take us maybe back to like Bio-CCC general because I feel like oftentimes when you when you look at either the legislative proposal or the impact assessments, it may still seem like you know there is Bio-CCS which is a large part of every mobile and then there's DAX. But then even there are parts in the equal assessment that specifically separates bio-CCS between power generation industry and biomass processing. And now we heard from ever how different the price range can be depending on the purity of the captured CO2 and so on. I'm just wondering ever if you look at the potential in Europe, we know that the commission would like to be as cost-efficient as possible when they're allocating this funding for remobiles. And if cost efficiency is the first thing, then obviously you know the cheapest capture options will likely be the first ones we get that get funded. So I'm just curious like how much of it would be available do you think. And also the challenge of what would we want to see happen to actually scale up the sector and how we can see this proposal like turning us how So it may happen because we would want to have different types of projects being supported. We would want to have geographical distribution and so on, but just because the effectiveness necessarily doesn't take us there. So what are your thoughts? I mean, like cost effectiveness is obviously a very important factor to bear in mind. But I think, again, to kind of bring it back to what I said at the beginning, we really need to focus on the bigger picture. So this isn't ultimately an exercise in finding what's the most cost efficient form of CDR. I mean, at least in Europe, we can agree that the CRCF is a pretty robust set of standards that we have for doing high quality carbon removal. But so within that, then we also want to make sure that we're not overly reliant on one particular technology. And I'm sure also there will be qualitative factors added to the purchasing program that the commission will roll out to ensure some sort of geographical diversity. And I think that that's certainly a given. But I'm not trying to dismiss this exercise in general of wanting to discover what the today, like what the price of carbon removal is. But again, the flat answer is, unfortunately, we simply don't know. There's not enough projects out there with real world data having done the actual costings for all of the materials that you would need to do an industrial scale project. I mean, we have a few first of a kind that are going ahead of different sorts and shapes and sizes. So that's good. And I think that's what the innovation fund is also there for is to also support those projects in not just discovering whether this technology can work at commercial scale, but also at least for the commission as well to understand how much approximately all of this would cost for a project to start out with. And then we've seen it from project developers many times. And I think also the folks at oxy and 1.5 have shown this that in their project they're building in Texas, which is obviously a very, very large direct air capture plant. They've built it in significant shorter amount of time on the first part of the unit versus the second part of the project. And that's just normal, you know, like the first time you do something is usually a little bit clunkier. It's a little bit more difficult. And then the second time you figure out, OK, here's where we could have made improvements. And then by the time you get to the fourth or fifth or sixth time, yeah, you've probably got things going and you've got the same people who've probably done the first, you know, second, third iteration of the project. And you can afford to see reductions there, certainly on your capex and probably also on your opics to some extent. But again, it depends on what we're talking about when it terms of individual projects and above all, what the ultimate goal here is. And I think that's something really that is a little bit ambiguous in the proposal when it comes to what exactly the commission is trying to achieve with the 250 minute on purchases. Two elements on bio CCS. One is the energy revenues of bioenergy systems. But I'd like to start with another one, which is technology diversity in banks. You mentioned like bio gas plants where CO2 is a co product that is kind of free of cost. And if it's close to the pipeline, yes, that can be cheap. But the volume is relatively small of these plants. But then on energy, there's one very important aspect. So when we talk about point source like x.xg like projects, I find it kind of difficult to imagine that many new plants will be built to provide heat to cities because there's very limited demand of heat of like 100, 200, 300 megawatt at a site. So having smaller units is much better suited to serve the heat market as compared to very large installations, large point sources. And then on the energy as well, the usage of or the process itself is highly energy consuming. So that's kind of taking away 40% of the residual heat that is produced by a large power of wine. So that can also be quite a significant effect. Excellent. And maybe to wrap us up here, I just did one for each section to give kind of the hard data I did find on today's prices because I do think that's useful. So in impact assessment, 2030 is 167 to 261 euros. And CDI for AI data, global today is 189 euros, which is actually below this. For Europe, it's 280. So there's almost 50% markup for Europe. And it would sit, I guess, at the high end of this. I actually also put some data on biocCS transactions from come future. And that is 220 to 340 euros per ton in Europe today. What's really important to note with all of the CDI for AI data and come future data, that has subsidies baked in, right? So this is a very important point to note. There might be a good 150, maybe 200 euros of subsidies baked into these prices. But yeah, I thought I'd share that. And now we'll move on to DAX. On DAX, there's actually not a single number we can expect, sadly. It's not like with biocCS, where I could tell you, you know, they assume X for transport and storage and X for energy. It all comes from the McKinsey report that I mentioned earlier. And the McKinsey report is a black box, sadly. The only thing we know from it is that they interviewed people from Climbworks, 1.5 and Carbon Direct in 2023. So we can assume that a lot of this data comes from projected costs in Iceland and Texas. But we don't have any specific heart, essentially, numbers we can work against. So they took these numbers directly from McKinsey. The data is global. It assumes falling energy costs. And this is an interesting contradiction I find with biocCS, because biocCS assumes very high energy revenue and the DAX focus assumes falling energy costs. And then, essentially, and this happens very often in these reports or modeling, there's just kind of a flat learning curve deduction for each year, right? The roughly 6% until 2035 and roughly 1% to 2% after 2035. So we just assume prices are going to go down. There's not a lot of explanations as to why. So my key questions here is, again, one, the energy costs, two, how representatives, the projects from Texas and Iceland would be for the European context, because that's what we're working with today. And then, yeah, I tried to dig up some costs from today. But yeah, ever it's over to you, again, what to take. And don't ask me how much my trousers cost. I won't tell you this. Oh, okay. That's damn it, though. I was lying that one up. But I mean, I think, look, with these reports, again, with respect to all the researchers that are there and kind of applying a flat learning rate or learning curve, let's say, you know, there is very few projects. That are out there. And I think your question of sort of how much can we learn from Stratus in Texas or Orkan and Mammoth in Iceland. Some things, I think, in terms of construction, in terms of how to actually cite a DAC plant, all the different contracts that are need to be in place. There are things that are, I suppose, you know, we wouldn't think about initially. But those are the, a lot of things that you can learn from it. Irrespective of the energy mix, obviously in Texas, you know, natural gases, a lot cheaper than it is in Europe. And also Iceland is blessed with geothermal energy. But I think if we only focus on the energy part of it, I think we're missing the point. And that's not to belittle the energy point itself. It's just, it's riddled with uncertainties. And I think that is kind of the point when we're talking about whether it's DACs or by a CCS or any carbon capture and storage project as well, is that, you know, in many cases, these are projects that are relying on infrastructure to be available, of the availability of that infrastructure being available is integral to their business case. So if a compressor goes down, if a pipeline isn't in operation or is available, or if there's also a significant time gap between the infrastructure availability and coming online versus their capture plant, you know, all of those things are risks. And I think we haven't talked about yet is sort of the de-risking element of these projects. In many cases, these are first of a kind of projects. And, you know, you mentioned the sort of 200 euro plus prices that may also have subsidies baked into them, that is also reflective of a project that's very, very risky. And so with those risks, investors will, I suppose, rightly demand a higher return. That's the kind of the nature of a high-risk investment. And I think overall, in the future, I think with better policy that's focused as well on the infrastructure side in trying to de-risk projects, that will also enable costs to come down and the risk reward ratio will be reprised. So that's, again, it's a bit of a long-winded kind of non-answer, you might say. But if you're asking me or asking anyone how much will DAC cost in Europe, anyone giving a number even, I think, is being a bit facetious if I'm in my, in my opinion. What I would like to add here is a comment that Martin Freimler, the CEO for all of Octavia, DAC company, didn't want one of my postage very much resonates with what you said, which was basically, I would heavily discount the word of any DAC's developer who hasn't actually delivered carbon credits yet on this question, including myself. Most industry outsiders and even insiders base their forecasts on wildly simplistic, bright and economic models disconnected from the messy reality of delivering DAC's in the field. Chair, comment. I thought that was worth a readout because, yeah, that sums it up. On costs, again, I just want to finish up with kind of the stuff that I did find. The impact assessment in 2030, so let's start with the kind of today version almost, assumes 288 to 576, Europe's, CDI for AI gives us 561 for European DAC's, so it's kind of pretty much exactly at the high end of this range and 445 for global, so there's a 26% discount on global DAC's compared to European. We see this discount across technologies, by the way, as I said, for by CCS it was almost 50% here, it's 26. Let's move on to the last category, biocha, biocha common removal. This one is particularly interesting because the low cost of the modeling is used explicitly in kind of the rationale for not including biocha. So that's why it's the only one of the three where the cost actually referenced in an argument like that. The low bound numbers, which are very low, 66 to 37 euros per tonne. Come directly from, again, the McKinsey study, which is a black box and is very hard to decipher. But I've done a lot of work on similar estimates for biocha. And when looking at it, usually they're pretty clear assumptions. It's usually a seasonal biocha, not industrial biocha. They're usually global numbers. There's bigger assumptions on how much people would get for their biocha sales, for the physical product, how much they might get from energy revenues that they might get free biomass. So there's a lot of unrealistic assumptions that often play together to reach this number. And again, global attisinal biocha. So this number is very low, which, again, pushes the whole median very low. The high number, this is also fascinating. This comes from Rambo, who I mentioned before, and they did a public survey that more than 100 people took part. And again, random people, anyone could participate. A coup d'angio, an individual, a corporate, could be anyone. Of these 100-plus responses, eight ticked the biocha box and gave an estimate of how much biocha could cost, 2030, 2040, 2050, and so on. And essentially, these ranges were like 100 euros to 400-plus euros. And what the commission did is they took the median from the survey as the high-cost estimate, which is not logical, because the median shouldn't be the high-cost estimate, but leaving that aside more importantly, eight random people filled out the survey. We don't know if they have any relation to the biocha industry, for example. And Rambo, to their credit, was pretty clear in their study that this data is maybe directionally interesting, but it shouldn't be relied on too much. So this is the high-cost estimate, which I thought was, yeah, rather strange. But yeah, so this is how you get a very, very low-cost estimate, a still very low high-cost estimate. And then the median I already explained is always split right in the middle. And importantly, none of this data is inspectable. Both the McKinsey and the Rambo data, we have no idea really where it comes from who provided it and so on. Now, Hansurk, I wanted to understand a couple of things here. One, again, how relevant is this global data that forms the lower bound, like the McKinsey 66 to 37 euros? And yeah, how do you, I mean, again, you've been compiling this data every year for the last couple of years. What was your take when you read through it? - I mean, first of all, global south data is irrelevant to European standards. We have to distinguish that we have severe emission regulations here and a total different cost structure on personal cost on biomass. So it just doesn't make sense to take global data. But let me start with saying, where are we today on the biochar industry? So as you know, I'm doing the European market report now for the fifth, had done it for the fifth time now. By end of 2026, we have around 235 biochar plants in Europe. We have roughly 30 equipment manufacturers in Europe. Around half a dozen is TRL 9. So TRL 9, there's more than five companies that are TRL 9, that are systems that deployed several industrial systems. Europe is leading in terms of technology when it comes to efficiency, when it comes to meeting the most severe emission regulations and when it comes to biochar quality. And biochar quality has a lot to do with permanence. So if biochar quality is well determined, then there's just no question about the permanence of this biochar. I think it is important to mention that this is an industry where there is real traction. And now I come to the cost, because we as biochar industry, we know exactly what the numbers are. And the lower bound is around 150 euros, where you can deploy carbon credits from biochar in Europe. But these are rather special cases, where you have either high revenue from the black thing. So when you do biochar business, you always have the black thing and the CO2 element. And if you have an application and you have a customer that pays a lot for the black, then you obviously need less money for the CO2. And there are niche markets where you get a lot of money for the black. And then you don't need that much for the CO2. But the scalable, the real scalable business happens between 150 and 250, or let's say 175 and 200. But that will only happen if there is a remuneration scheme for the carbon credits in the range of, let's say, 150 to 250 euros, or to narrow it a bit down 175 to 200 euros. Then it scales. Again, to give some of the data that was out there, the impact assessment assumes in 2030 66 to 250 euros. I have data from CDI for AI, which for the EU, there's an average transaction of 192. Global, it's actually 116. So there's a 35% discount. Then I have data from Rainbow. They actually commented it on my post. Thank you very much. They said that they have 160 to 300 euros per ton, average transaction data, and also looked up the COM future data, which was 130 to 300 euros per ton. So I think there's a pretty clear picture on at least where we stand today. So here we have it. We've gone through the three different technologies in the proposal. We've checked some of the assumptions. We've shared some of the current real costs and also spoken about potential of those technologies. I wanted to finish with three key asks that will apply to anyone listening, basically, in some way. The first one is for people who are doing these modeling exercises, please do use CDI for AI data, up to date CDI for AI data, wherever you can. They've got a portal you can log in. That data would have been more accurate than a lot of the data we actually discussed today. So please go there and check out their portal. It's worth it. The second one is be transparent about data sources, gaps, shortcomings, assumptions. I thought that Ramble did this really well about their survey, but actually, usually it takes a lot of digging to actually understand how these numbers come together. A simple site sentence explaining some of the assumptions could go a long way, in my view. And the third one, and probably the most important one, is supply your data to CDI for AI. Whether you are buyer and intermediary or especially a supplier, you can supply your data. You can even tick a box to say that it should never be published by anyone, so it's just for CDI for AI. But that is the most obvious way to create a shared knowledge base of where we're currently at. So yeah, there's way to little data submitted to that portal. So please, please submit your data. And with that, I'll thank our two guests, Ever and Hans York. Thank you for coming today and for sharing your knowledge. And yeah, have a good rest of the evening. Yeah, thanks so much. Thank you. You're welcome. It was great to be here. Thank you. Bye. Tina.

Podcast Summary

Key Points:

  1. The EU Commission’s carbon removal cost estimates for bio-CCS and biochar are primarily derived from four sources—McKinsey (2023), Ramble (2025), NEGAM (2023), and CDI for AI (2023)—with significant reliance on outdated or incomplete data.
  2. Bio-CCS cost assumptions are heavily influenced by energy revenue, with low-cost scenarios relying on high energy prices and negligible transport/storage costs (e.g., 38 euros/ton), which are not representative of real-world European projects.
  3. Real-world project data shows bio-CCS and biochar costs in Europe are significantly higher than modeled, with actual transaction prices ranging from 150 to 340 euros/ton, and biomass sourcing being a major cost driver.
  4. Direct air capture (DAC) costs are poorly modeled due to reliance on global, non-EU data (e.g., Texas and Iceland), and lack of transparency on assumptions, energy costs, or local applicability.
  5. The Commission’s low-cost biochar estimates (e.g., 66–37 euros/ton) are based on global, non-European data and unrealistic assumptions, while actual European market data shows costs between 150–250 euros/ton.
  6. The proposal assumes cost parity between carbon removal and emission allowances, but real-world project economics, risks, and infrastructure constraints suggest significant overestimation of affordability.
  7. Key data gaps include lack of transparency in methodology, absence of real-world project cost data, and underrepresentation of supply-side data in public models.
  8. Experts recommend improved transparency, use of up-to-date CDI for AI data, and voluntary data sharing to build a credible, real-world-based carbon removal cost database.

Summary:

The EU Commission’s carbon removal cost estimates in its ETS review rely heavily on a limited set of sources—McKinsey, NEGAM, CDI for AI, and Ramble—many of which use outdated, global, or speculative data. For bio-CCS, costs are modeled at unrealistically low levels due to assumptions about energy revenue and transport, ignoring real-world biomass sourcing and infrastructure costs. Project developers in Europe observe actual costs ranging from 150 to 340 euros per ton, with significant variation based on project scale, location, and feedstock.

Direct air capture (DAC) costs are similarly uncertain, relying on non-EU case studies and flat learning curves without justification. Biochar costs are similarly skewed downward, with global data ignored in favor of European-specific realities. Experts emphasize that current models fail to reflect real project economics, supply constraints, and infrastructure risks.

A critical gap is the lack of transparency and real-world data. To improve credibility, the Commission should adopt transparent, up-to-date data from CDI for AI and encourage all stakeholders—including buyers, suppliers, and developers—to contribute real project cost data. This would enable more accurate, realistic, and actionable carbon removal cost modeling for EU policy.

FAQs

The data primarily comes from four sources: a McKinsey 2023 report, a Ramble Ecologic study from 2025, the NEGAM study from 2023, and CDI for AI data from 2023. These sources are used to build cost scenarios for bio-CCS, DAX, and biochar technologies.

The assumptions rely heavily on limited, often outdated or non-transparent data. For example, energy revenue assumptions are based on unverified claims from companies like Drax, and transport/storage costs are unrealistically low, with no real-world validation from operational projects.

Real-world data from Europe shows current bio-CCS costs ranging from 220 to 340 euros per ton, significantly higher than the low-end estimates (172 euros) in the impact assessment, which also include significant subsidies and assumptions not reflected in actual operations.

The impact assessment estimates DAX costs between 288 and 576 euros per ton in Europe by 2030. CDI for AI data shows European DAX costs at around 561 euros per ton, indicating high prices, especially compared to global rates which are discounted by 26%.

The assessment explicitly uses the low-cost estimate (66–37 euros per ton) from McKinsey to argue biochar is cheap, which is a significant oversimplification. This low bound is based on global, non-European data and unrealistic assumptions about biomass and market prices, making it misleading for the European context.

Real-world data shows European biochar costs range from 150 to 250 euros per ton, with scalable, viable operations typically falling between 175 and 200 euros per ton, especially when supported by a carbon credit market in that price range.

Chat with AI

Loading...

Pro features

Go deeper with this episode

Unlock creator-grade tools that turn any transcript into show notes and subtitle files.