Germany’s Battery Deadline Comes Before the Grid Rules Do - Kyon Energy

I'm your host, Ed Porter. Welcome back to Transmission. An end-of-year deadline to manage a 20-year exposure. That's the deal on the table for German battery developers. Reach a final investment decision by the 31st of December. Commit to 50% of the required component orders and get the battery live before August 2029. And you pay no capacity-based grid fees for 20 years. Miss the cut-off by a day and the fees start in 2029. Then there's the catch: the rules on how those batteries can actually use the grid, from dynamic fees to flexible connection agreements that won't be settled until after the money is committed. So how do you commit to a battery when you can't see the terms it will run under? Marie-Sophie Braun is Head of Markets and Regulation at Kyon Energy, taking projects from development into operation. She's making those calls right now. And if you want to see what German batteries are earning today, or how that compares with solar or wind, ask Ko, Modo Energy's AI analyst. Let's jump in. Hello, Marie, welcome to Transmission. Thank you very much for having me here. Our pleasure. And as ever, let's get straight into it. So what is the one thing that people get wrong about the German battery storage market right now? I do think to be honest, during the last year the discussion was still centred about around this battery tsunami. So this really high a number of grid requests. I do think it's it's not a good measurement in general on how storage deployment will be in Germany. It might be a sign that there is still interest in in the German market, which of course is and I think for for a project developer and operator, that apparently people believe that there is a business case, but in the end it has no impact on the number of projects that will be deployed, will be built, and will be ultimately also integrated into our energy system in Germany. Yeah. And it's kind of fascinating, right? So that when you talk about the battery tsunami, right, the number of grid connections that came forward, that was like around 300GW or so, more or less. I don't I don't don't quote me on that number, but it's kind of more or less that number. And when, when we do the the forecasting for how many get built in Germany, it depends exactly what, what date we're covering. So like whether it's by 2040, 2050, 2060 like whatever it might be, but the number is probably somewhere between like 50 and 70GW, right. So if you're going from 300 down to 70, you've kind of you've got a lot of applications coming in. But yeah, as you say, like if you're the 301st gigawatt that joins the back of the queue, you're probably not going to get built. And it's also not in the interest of the storage industry, to be honest, like it's a merchant asset that needs to to earn back its money. So also from the industry itself, there is no incentive to have 300, 600 gigawatt of storage installed in Germany. Yeah, and sort of unlike other sort of unlike some other energy assets like storages, generally competes on just a pure sort of supply and demand basis. Right. So the thing you're alluding to there is you're saying that as you add more batteries to the system, they compete with each other and you get this kind of compression of margins, which means that the batteries that have been built all suffer if lots of overbuilt happens, which is kind of a fascinating thing. Okay, I'm excited to get into German BESS, and maybe we'll downgrade the tsunami to a wave as the conversation goes on. Um, and let's talk about Kyon. So you started life as a developer and you've moved into sort of operating and optimising your own assets. It's kind of an interesting transition and not everyone does it. So why did you decide to hold on to those assets and operate them rather than what some developers do, which is you kind of get the rights and then just sell that off to another party. I do think Kyon, as a as a company, always had a very clear mission and also somehow a kind of ambitious like vision for the storage industry in Germany. So we really came into the market and wanted to shape the regulatory framework. And there were maybe three kind of motifs that still are true today, to be honest. The one would be that we believe that storage in itself, it's not just an asset like a customer and something like a producer of energy, but really its own asset class. We do believe that the last few milestones of the energy transition in Germany will only be feasible. And yeah, also affordable. To be honest, when we integrate flexibility somehow efficiently in Germany, and we do believe that this is still possible without having a big financing scheme from some from from the German state. And I do believe that this transition from being just a project developer to an operator is kind of delivering that mission as well for us. So we kind of now are deploying this vision that we have been telling people since a few years and call ourselves now an independent flexibility provider. So, so that's that's good for us. And I would say this has always been the purpose of Kyon. And for me, one of the interesting parts about that, right, is that as a business, you'll have built up this ability to find sites, get grid connections, acquire the land, and then finance that and bring it through. And there's a particular set of skills that exist there. And it's sort of the classic developer set of skills. And the next set of skills is how do you best optimise this? How do you best operate these assets? Sometimes not the same set of skills, right? So as a company, have you had to can you can you kind of see that the company is changing a little bit around making sure that it's got that set of skills to deliver that next part? Of course. I mean, Kyon had to grow up together with the storage industry, and I do think that's a huge advantage that we have in-house because, for example, we had the chance to start with the smaller projects. Uh, it's it's better to learn on a mid voltage project, I assume, than aiming high for the, the bigger projects right away, just in order to kind of understand the processes, learn from it, have this like fast learning organisation that is able to deploy as an asset, learn from it. And then when the next greenfield development starts, we kind of already had this full cycle once, once done in-house, which enabled us to be much more efficient for the for the next project. But obviously there has been a lot of new learning experience. We're building an asset management, um, which was obviously was not there before. And this requires a very high skill set. There's people like, uh, like the two regulation cybersecurity rules that need to be respected. This was out of scope before and still challenging, but but good to for us as an organisation to to learn from it. Okay. And it's interesting. We had an episode recently with Alexandra and Jan from Entrix, who were talking about Poland, and they were saying that because they haven't had the ability in Poland to sort of add in these 1 to 5 to ten megawatt sites, and they've had to go in with a 200 megawatt site, essentially. So they've started with a very large site. They've missed a lot of those stages that you've just talked about, which is like we learn on this asset, we learn on this asset. We got better at these these pieces. So it's interesting to see different countries do this in different ways. I think the, the, the approach of kind of allowing those smaller projects to come through is really important because it allows groups like yourself who are tied to the battery industry to to build up the capability that you want to. And maybe one sort of final question on this. So, um, one thing we see quite a lot in, in independent producers, uh, or flexibility providers, however you want to sort of describe them. Like there comes a question of like, how much do they bring in-house? Because there are loads of good optimisers out there, but there's always this kind of temptation to to do it in-house and to sort of become an optimiser yourself. Do you see that temptation, or is it something that, um. Is it something where, like there's you're pretty happy with the route to market you've got? I do think this is a whole different skill set, to be honest. Um, because also our interest is really asset centred. And of course, you know, we want to educate ourselves. We want to be smart about the market option. We want to understand what's actually happening beyond just the euro per kilowatt hours. And we want to learn how the asset interacts with the grid. Okay. Um, but still, I do feel like there is a lot of really good solutions already on the market, and it would not be worth it to internalise it when we can also have very strong partners alongside of us. Yeah, it's really interesting. It's like a it's like a global like, um, point of contention for developers and owners. Um, but let's, let's kind of move it back to the decisions that you're taking right now. So, um, you're being asked to take final investment decisions this year on assets. And those assets, their grid usage terms won't be settled until after you've committed to those projects. So perhaps two things. One, could you kind of describe a little bit of what you're seeing in those grid usage terms. And two, how do you as a company, how do you make that decision? To be honest, it's it's a big sprint for us for now. And I mean, the the background being the reform of the grid fees. It's called AgNes in, in in German. Uh, this was something that was, is actually a Germany first that the German regulator is the, the party who's taking care of how the grid or grid fees in Germany for all asset types should look like it's a huge task for them as well. It has a lot of impact on more industries than just the storage industry, but for us, of course, it's really important because for now we are exempted from from grid fees. If your asset goes live before August 2029 and this grid fee exemption is not going to subsist if you don't take FID by the end of the year, at least that's that's the the current knowledge base, I would say. And FID is not just a signed paper, but really you need to allocate a big portion of your CapEx for the asset still this year. Um, and for us at least, it's it's really ambitious to do so just because of the tight timeline we know of this deadline, since I would say May of this year. And it's like if, if you really want to do infrastructure investment. Seven month is usually not enough. Um to close financing to to do your due diligence on supply agreements. We do think we do have a fair chance to to achieve that. But it's now much more about active risk management because we have this deadline. By the way, I mean the business case still will subsist. So it's not a harsh deadline. And then no battery buildout in Germany is going to continue. So your project next year like like how are you going to think about those. I mean better to have the grid fee exemption. But I do think that's my personal opinion that the regulator has found a good compromise for everyone now, like storage will pay the same grid fees as, uh, also renewable assets, for example, will will pay. And this is something which is understandable in a situation where the bill that has to be paid for grid build-out is getting bigger and bigger, and this can not only be given to consumers. Yes. So I think it's it's a fair regime like the the business case of course is better when you don't pay grid fees, but it will subsist. And hence it's a commercial decision to take FID this year. And it's a really ambitious one because in the meantime, in Germany, we do have a big discussion on how are we actually going to integrate all these storages into the grid while keeping our security of supply. And I think there's some really interesting parts of those, those grid fees, the one that I, uh, from my reading, have liked the most is the the sort of look at locational and dynamic grid fees. I think it's like a really nice concept where where you're showing that you've got congestion, you're encouraging the storage assets to act in a way that is going to help to resolve that congestion. That to me feels like. Sorry. I borrowed something you said you were kind of saying. It's. It's kind of like a reflective way of sort of treating the costs. And it seems that sort of seems more fair, um, to me, like a dynamic and locational grid fee. It feels like it does that. Right. And it encourages assets to work in ways that help the grids rather than just sort of layering in extra cost. Mhm. Like is that your take. Yeah I would agree totally. I do think of course you know like every model you will build that's never going to be perfect. There will be the point in time where you've set the grid fee maybe one day in advance. And then there's an unexpected event and you will have a wrong incentive. That grid fees is only part of the solution, or dynamic grid fees are only part of the solution. It will be still a combination of redispatch and other measures. And. But I also do believe, to be honest, that it's really important to incentivise storage to act on grid needs because for now, to be honest, our storages are kind of blind when it comes to to grid congestion. And for people who are listening to this who aren't necessarily in the detail of the German market, redispatch, especially when you get close to delivery, the assets are really dispatched to account for things like congestion in the system. Exactly. That's that's the normal process in Germany. It's also hard to include storage into it because the processes are not fully there yet. But we see it more and more and obviously that's a good thing. Storage should be integrated as every other asset into German congestion management. Yeah, yeah, I totally agree. Um, okay. Moving on from those decisions around AgNes changes, uh, do you think that there is a and this can be both be for sort of AgNes. So from the grid connection side, but it also can be from sort of flexible connection agreements as well. What sort of splits a regime that a developer like you can support and actually finance, versus one that sort of stops everything. As a storage industry, we have always been used to market risks, and that's that's the game you're playing when you try to deploy a storage asset. So I would say we we are inherently kind of good to manage uncertainty when it comes to these FCAs. We do see that there is a few ones which are hitting very hard. This particularly comes into play when, uh, you're faced to congestion management management which is not remunerated. For example, there is there's schemes which are discussed which just do not limit the capabilities of your connecting grid operator to just curtail you without reimbursing you. And this is something which of course, I mean, you can do the scenario analysis. You can see with different weather years how it's going to be. Yeah. but the risk profile just increases so much that it will be really, really hard to to find financing for not only one asset, but for a larger portion of flexibility. And this is something that I feel should be limited. And it's an idea that has been floating around at Kyon since 2024 or 3 even, I think to kind of grant the grid operator a certain portion of non-compensated redispatch measures somehow. This is something that we can do if it helps to alleviate the grid, but it needs to be a really like distinct amount of numbers. And I think also from other market participants for example Green Flexibility. You can see that this is something which is achievable through negotiation with your grid operator. Did you did you catch the Green Flexibility episode on Transmission? A catch. There. You good. Very very good, very good. It's a great episode. Um, uh, but yeah, I totally agree with you, right? So I imagine that I'm a developer and I'm going to a financier and I say, hey, um, can you can you put debt into my projects? And they say, oh, yeah, sure. Like we can do that. And he said, well, look, I've got this connection, but I might get turned off for the year. Like no financier is going to say yes to that. So as you say, like you've got to have the you have to have like a very clearly defined like this is where my assets are going to be available or else bringing debt into that project is going to be super tough. Well, let's move from the theory into like real life of, uh, you have an asset coming online soon. I don't want to put a date on it, but you can you can do that for me. Um, and I think you're going to split the asset to have three different route-to-market providers. Uh, I believe that might be correct. You can correct me in a second, but like, by by splitting assets out across different route-to-market providers. Um, what are you trying to learn with that approach? Well, what we do is actually we we as Kyon, we do have a like a bit bigger portfolio of storage assets. And we do not split each asset, but we have comparable assets and we give them to different market providers. Of course, that's still a little bit. Not comparing apples to apples, but kind of ish. And what we really hope is of course, you know first it's a benchmarking approach. The best market is still emerging. There are a few established ish players from from the route-to-market providers. But I do think we want to see and learn from for ourselves. Beyond the shiny PowerPoint presentations, who's actually performing on live assets? And on the other hand. We also are seeking a long term partnership with our route-to-market providers. We do see, and I think that's that's not a secret that the regulatory environment is changing fairly quickly in Germany. And for us it's a strain on the business case. But for the route-to-market provider, this is something that he will need to act on in real life. So when a load gradient, an FCA regime comes in, it's the route-to-market provider, which is the responsible party to really deliver on these new rules. And this is something where you need really well oriented partner to achieve that and not have not having downtimes on your asset, for example, because they're late on algorithmic development. That's exactly what I was going to say actually. So we we on on Transmission, we've interviewed a lot of of of optimisers and occasionally someone will come in and say, ah, we're better than all the other optimisers. I really don't hear that very much these days. Actually, I probably more often than not, I hear we help parties get assets online. We're up to speed with the regulation. When there's something that goes wrong on the site, we're quite quick to get it back online, because I think what they're pointing at is that the way you can really lose money on assets is like being a month late or being, you know, uh, just just missing some element of the dispatch that you should be running. So actually like the physical delivery of their of their performance is kind of more important than having a slightly more refined trading algorithm. Maybe an optimiser will be listening to this and be like, hey, you're so wrong, I'm going to come on and tell you why you're wrong. And I look forward to that day because I'm sure it will happen again soon. But, uh, yeah, that's definitely the themes. Those are the themes that we're hearing and, and in terms of different assets. Right. So you mentioned lots of different assets coming through. And there's obviously transmission and distribution networks. Those networks have different incentives from their grid operator. Do those networks sort of change where you'd rather build as a developer. I do believe it's worth when when I would start a greenfield development today. Uh, I think the biggest work is to really investigate your grid connection point. Um, because also kind of as an asset operator, you might even be able to manage your expectation what the result is going to be this. There can be a lot of factors to that. For example, if you know, and these datas are partially public, that the grid area is highly congested already, I mean your chances are getting lower obviously. And that's a natural thing. On the other hand, of course we do see a difference between a TSO and a DSO connected assets. And a lot of people are not really happy about the DSOs. I do think that's not fully true. We do see really like highly engaged DSOs. We do see highly digitalised DSOs. We do see DSOs that we are really like working on integrating storage into their grids fairly efficiently. Okay. So so that's that, that might be the olive branch there in the in the same time of course, just from a from the way they are regulated, there are a few incentives which align much better between a TSO and a storage operator than it is the case for a DSO and a storage operator, like giving you an example. Ancillary services in Germany are provided for frequency control, and obviously all grids need a stable frequency, but it's just not the job of a DSO to care for a frequency or for the stability of the frequency at least. So when you provide an ancillary services, be it FCR, be it maybe even inertia, and this is a service that you provide to the TSO and might create issues in your DSO network. This can create voltage issues, for example in the DSO network. And the DSO will only see the trouble you're making and not the benefit, because it's just out of the scope to care for frequency control. Yeah. And this is an issue which needs some kind of regulation. To be honest, I do think like the only party will who is able to kind of decide what is more important will be the regulator, because obviously the TSO wants to have the ancillary services. We want to make money on the ancillary services. The DSO has has no like, proper incentive beyond being happy about storage, deployment, maybe to see it in in its grid. So that's that's an issue which creates a lot of friction but is solvable through proper stakeholder management explanation like proper like discussion on a technical level to really understand the issues your connecting grid operator has or might have with your asset. I think this is a fascinating part of the market. And um, the conversation just kind of going beyond storage, just, um, uh, charging of the low price and discharging at a high price, which is kind of what people think about it most simply to go into things like voltage control, to go into things like inertia, a frequency response and frequency response is perhaps more and more well understood. Things like fault current being injected into the system. I think this is where the conversation is going. Over the next 3 to 5 years, we'll find the umbrella term of system strength and all the stuff that goes underneath it. I think we're going to start to think a lot more about and like be more tactical about where we place batteries onto these systems. And maybe this kind of ties actually really into into a nice sort of market design question. And one is everywhere. So this, this kind of same thing is happening right where we've gone from really large transmission, connected big power stations to all of a sudden solar, wind batteries, much more distributed assets, they can kind of be placed anywhere. Um, and what's happening is that the, the big sort of national system that we used to have is being questioned and saying, well, actually, do we want a big national system or should we have a zonal system or should we have a nodal system? And Germany is going through that process. Right. There was a few. This was looked at a little bit this year. It was also looked at a little bit last year as well. Where does Germany stand on the system design? Where do you think it might go? I think there there are more and more actors really rooting for a split in Germany. That's, that's that's for for sure. True. Is that is that. That would be zonal. That would be zonal. Yes. Um, that's I mean, it's it's an ongoing debate, I think, like the congestion issues having a lot of energy production in the north of Germany, a lot of customers in the south of Germany, that's that's a problem that has been recognised. There is a lot of like DC high, high voltage cables being built out in Germany to kind of try to cope with that. Uh, but we see that now it's not enough because just the, The the amount of money we spend each year on redispatching, which is kind of realigning the price. The the price signals with the actual reality of the physical energy network, um, are getting higher and higher. I do think from a flexibility perspective, of course, uh, it's an advantage to have to have a to, to have price zones because then assets can actually react on more local signals. Yes, that's that's the big issue. I mean, I've been talking about it like it's a pity that there is no real mechanism where storage can are enabled to align with the grid needs beyond an FCA regime. And the dynamic, uh, grid fees. Yeah, yeah. Coming in, that will be coming. But for now, like for now, there is nothing beyond like, your connecting grid operator telling you you're just not allowed doing certain time windows to to act freely in in the grid. And it's a pretty it's a pretty bad solution for consumers. Right. So consumers are told oh by the way, you're going to pay large redispatch fees. And also we're going to like the lever that we've got to pull is this big HVDC line we can build from north to south. It's not it's not really great. Yeah. In the in the same time, you know for me this and this is probably also where it hasn't been solved so far. It's an issue that goes beyond the storage industry and maybe also beyond the energy industry. Because when you do it, this will have an impact on other industries, on people living maybe in certain areas where electricity prices go up. So when talking about this, I think you really need also to assess the social factors that come with it beyond also operational complexity to maybe cut a German TSO in half. Mhm. Yeah. Yeah. It's it's going to be a it's going to be an interesting, interesting debate right. And it rumbles on. Also for people listening outside of Germany. It's not just a German thing, right? We're seeing this everywhere, where people are questioning the sort of single national market design. Right? Let me take me through to the final question. So what is a contrarian view you hold about energy in Germany or the battery storage industry in Germany? I would say like we do see a lot of good approaches also when it comes to reflection on grid usage conditions. Unfortunately, I see them more in the renewable sector. What what we see in the in the EEG, the German law for renewables, for example, that they have this idea on in highly stressed areas you will get congested without renewables, without remuneration, but only for a certain amount of hours and only for I think it's six years. And this regime makes a lot of sense. You know, there I can get behind it. It's something which is financeable, well, because we can quantify it. This would be an easy yes, I think for most storage operators, but this is something which is kind of discussed isolated for renewable assets. Then you have the big debate about grid usage conditions for storage assets, which is fairly different and mostly more restrictive. But in the same time, German regulation and also energy policy is kind of pushing for co-location co-located assets. And this is something in my mind that doesn't make any sense. Why would you introduce different rules for different asset types, which in the end are co-located? Because then just no one will know how it's actually working and which rule should apply in which moment. Yeah. So you're saying that, um, uh, the kind of status quo of the market is that you have this, uh, remuneration, uh, or support for renewable assets, uh, through the EEG around, say, congestion, but you don't have that for storage. And when you're starting to think about co-location as assets, the contrarian view is that we've kind of we've got this wrong, and we should be trying to align these two so that there's kind of a fair playing field between these, these two assets. Yeah. And also that the co-location just works from an operational level to be honest, because when you when you have yeah, when you just have different rules that apply to different to different asset types. Like it gets harder when they are co-located. Uh, which is something that is also pushed from the regulation. And we want it like we see it in, in the EEG regime. We see it through something which is called MiSpeL. That's actually a German regulation, which is defining metering concept, which enables you to do proper co-location, and also to switch between green energy being stored in your energy and your battery from a solar power plant, for example, and the battery just acting as a standalone base. But how should this work? Once you have done a dispatch policy that only applies to the green part of the battery, and another one that applies to the grey part of the battery. It's getting more and more complex. Instead of looking at the problem from the technical perspective, which is, is my grid actually capable of taking the energy or not? And when you're talking about green and grey, that's um, green is effectively a battery that is charging only from renewable. And grey is where it's also being able to charge from the grid. We don't we don't colour batteries differently. This is just how we talk about regulation as a in the sector. Okay. Thank you very much for coming on. You've been a wonderful guest. I think people will learn a lot about sort of how to think about the German battery space. It feels like so much is going to has changed over the last year and the next year coming. It feels like we're going to do the same thing again. But good luck with your projects coming online. I hope you are able to make those decisions and get them online. Thank you very much.