1&1 AG (1U1) Earnings Call Transcript & Summary

Good morning, ladies and gentlemen here on site. Welcome, ladies and gentlemen, in their webcast. I'd like to extend a warm welcome on behalf of our management Board of 1&1, Mr. Dommermuth and Mr. Huhn, CEO and CFO. The Management Board will present the results of the first half year of 2024 to you. Afterwards, we will talk about the forecast for the rest of the year. After the presentation, the management Board will be available to your questions here in the room. Unfortunately, no questions from the webcast will be possible, but we'll be happy to answer them throughout the rest of the day or tomorrow. Now I'd like to hand over to Mr. Dommermuth.

Thank you, Mr. Keil. Welcome, ladies and gentlemen. First half of the year 2024 and forecast for the rest of the year. I'll start with the company development and the status of our mobile network. Then Mr. Huhn will follow with the financial key figures and the forecast for the rest of the year. You know our business. We offer broadband connections and mobile networks. We have 3.99 million customers in our broadband network business of VDSL and FTTH complete package customers, including voice and IPTV. We purchase the products via our partner, 1&1 Versatel, and we usually use the network of Deutsche Telekom or city carriers for the last while, city carriers like NetCologne and others. So we cooperate with numerous different providers. Mobile, we have 12.36 million mobile contracts. We operate the first OpenRAN that is fully virtualized in Europe. We have a broad market coverage. Our primary brand is 1&1. We have co-brands with GMX and WEB.DE. And we also have discount brands in our group, like yourfone, smartmobil or winSIM. The first half of the year, we were able to win 90,000 new contracts, and we had a reduction of 20,000 in the first half of the year of our broadband connections. Revenue has increased by 1.1%. The service revenue is very important for us. We have a strong margin here. Here, we had a growth of 3.8%. We had some regulatory effects. If we take them out, we would have had a growth of 4.4%. Other revenue, that's specifically smartphones that we combine with contracts that will then be repaid throughout the length of the contraction, that decreased by 9.2% to EUR 371 million, and we have quite low margins in that area. Now if you look at our operating business. Access segment, we had an EBITDA growth of 11.4%. I think those are extremely strong numbers, 11.4% growth in our ongoing business. We started with our 1&1 Mobile Network. We started in December of last year. We're now providing mobile services as well, and we're in the process of expanding our mobile network. We had a loss of EUR 111 million, and EUR 14.3 million were out of period expenses. The reason for this is that we're preparing for massive capacities in our core data centers, in regional data centers and also when it comes to frequencies, everything that we need to gain further customers. And at this point in time, the network is not fully used yet. CapEx EUR 34 million in the first half of the year, significantly less than last year. The reason is a phasing effect. We will have higher expenses in the second half of the year. Specifically for our network expansion, we will have EUR 26.5 million for our mobile network. Now let's talk about the mobile network. I mentioned it. We have the first OpenRAN in Europe. It's fully virtualized, which means we use standard servers, and we have a mobile connection via software. Here, you can see our architecture. On the left-hand side, the connection to the Internet and other networks. This is all controlled by 4 core data centers. We currently have 2 core data centers that are operational, 2 are going to be operational soon. Then we have 24 decentralized data centers that are all fully operational. We're planning to have over 500 regional data centers in order to supply 50% of German households. That's our first target. In these data centers, we have standard servers running. It's a private cloud. And then antenna sites are connected to these data centers in order to supply these 50% of households. We will have 12,600 antenna locations. The antenna locations or antennas are controlled by the data centers. The locations or the antennas are, so to say, stupid. They cannot be controlled by themselves. All of this happens through the regional Far EDGE data centers. That's where the rated radio connections will be transferred into IP traffic or the other way around. The antenna locations are connected by a glass fiber. Every site has gigabit antenna. What's the benefit of our OpenRAN? It's an open system. We use standardized interfaces. We can use components of different partner companies, services of different partner companies. We currently work with almost 100 companies, and that makes us independent of other manufacturers. Our architecture is ready for real-time applications through our regional data centers. We cannot only control antennas. We can also get ready for applications, which means that the data doesn't have to go through the Internet but rather can be processed right on site. Our network also has lower energy consumption than conventional architecture. If we look at the antenna sites, we've made good progress. Many of you know that we struggled with this in the beginning. We wanted to have the first 1,000 sites by the end of 2022. Our most important supplier, Vantage Towers, has failed us. The antitrust authority is investigating in this matter, but still, we are able to make good progress in the numbers. You can see on the left-hand side, the number of antenna locations that we currently have. In the first quarter, we had a growth of roughly 280 sites. In the second quarter, we had a growth of roughly 450 sites. And this number will grow every quarter. So we will grow these antenna sites exponentially. On the right-hand side, you can see how far these sites have progressed. To start with, these are only masts on a property or on a roof, and the dark the dark blue area is antenna sites that already have an antenna. And then the lighter blue are stations that have antennas and glass fiber. We are currently at 456 fully equipped sites and 300 sites where we already have antennas but where glass fiber networks are still under construction. So you can see that we've made great progress. If you look at it quarter-by-quarter, you can see that in the first quarter, we've made bigger progress than in the first quarter, and we were able to equip far more antennas than in the first quarter, and this is how it will continue in the third and fourth quarter. So we're on track with this. Thank you, and now I'd like to ask Mr. Huhn to present the key financial figures to you.

Yes. Welcome, ladies and gentlemen. So let's start with the P&L for the first half of the year. On revenues, Mr. Dommermuth has already mentioned that you can see that we have increased it by 1.1% to EUR 2.015 billion. Cost of sales, a significant increase by 5% to EUR 1.458 billion. And that is due to the fact that it includes expenses which are used for constructing the mobile network. And that's why the gross profit had to be adjusted a bit. So you can see that it's decreasing from EUR 604 million to EUR 557 million. However, looking at the Access segment, you can see a very nice increase from EUR 659 million to EUR 716 million. So for the final customer business before expenses from mobile networks, we can see that the gross increase is 8.7% in our gross profits, and that really shows that also we need to look on the value that's behind it. So the gross profit from turnover for the 1&1 Mobile Network last year, EUR 54.6 million in expenses; in the first half of the year 2024, EUR 158 million. Distribution costs have increased slightly by 3.4% to EUR 262.9 million. Half of them are due to the fact that we had higher commissions, sales commissions, and the other half is due to higher overhead costs. Administration costs, EUR 57 million. That's the previous year level. And other operating income or expenses from EUR 13.5 million, they have increased to EUR 18.7 million, and that is driven by higher revenues from dunning and repayment fees. In the next item, we can see that there were higher expenses from impairment losses on receivables and contract assets, so minus EUR 52.4 million. They have increased to minus EUR 59.4 million. EUR 5 million of that is due to losses on receivables and EUR 2 million is due to higher impairments on contract assets in the context of IFRS 15 because that needs to be activated. That leads to EUR 196 million of profit loss from operating activities, over EUR 254 million in the previous year. Financial result first half of the year, EUR 0.4 million; in the previous year, plus EUR 3.4 million. This decrease is due mainly to higher interest payments from leasing liabilities according to IFRS 16 because we have to record the leasing contracts in a way that we have to indicate the interest expenses, and it needs to be shown up as if we activated those leasing agreements. Other than that, United Internet, actually, the figures are slightly below the previous year level, around EUR 5.8 million -- or EUR 8.9 million rather. So consolidated profit before taxes, EUR 257 million in the previous year, EUR 196 million this year. Tax expenses, minus EUR 60 million, and that leads us to a consolidated result for the first half of the year of EUR 136.4 million. Now let's have a look at the balance sheet. The balance sheet total increased from EUR 7.74 billion to EUR 7.92 billion. Let me just address the main changes. For example, short-term assets in the first half of the year, they were reduced by EUR 80 million based on the fact that we had reduction of investments with United Internet because the money was needed. So that's around EUR 100 million. Accounting for this item, also our reserves over the end of the business year 2023, were reduced by EUR 52 million. And we increased the deferred expenses by EUR 55 million, and that is due to the contract for the quotas we had because we had to make payments to telecoms. You will find that in the short-term assets, around EUR 55 million, but also in the long-term assets. And as you can see, this item increased by EUR 262 million over the end of the previous business year. The large share of this is due to the payment to Deutsche Telekom, EUR 180 million roughly; and further increases, around EUR 100 million, is due to investment assets. So first of all, whatever we invested but also the fact that according to IFRS 16, the leasing agreements, they had to be activated in the balance sheet. So let's look at the liabilities. Short-term liabilities, minus EUR 42 million; and also due to a decrease of trade liabilities and long-term liabilities, around EUR 100 million. That's what they have increased. That is because they are now recorded as a liability. So that is the leasing agreements for the antenna sites that are now recorded as liabilities. Here, you can see the cash flow statement. For the net inflow of funds from operating activities, we arrived at minus EUR 24.5 million for this first half of the year; previous year, minus EUR 23 million. And you can see the individual line items. I'm not going to go through them one by one, but maybe two things. Cash flow from operating activities, EUR 262 million this year; and the previous year, EUR 260 million, so this line is developing in a stable manner. And the change that contributed to minus EUR 24.5 million. That is mainly tax payments, and they are due to the fact that the prepayments that were determined by the financial tax office last year were made. However, the actual taxes incurred have been higher, and we had to make a payment, which we didn't have in the previous year, and that is the main change that contributed to minus EUR 24.5 million over the plus EUR 23 million in the previous year. Cash flow from investment activities, a positive result, EUR 69.5 million over minus EUR 0.8 million in the previous year. What's the reason for that? Well, this year, we invested less, EUR 34 million in CapEx. In the previous year, it was EUR 81 million. Then EUR 92.5 million, that's a reinvest of free cash with United Internet, so around EUR 90 million that were taken out and which are recorded as being positive. The same goes for interest, EUR 11.1 million, so interest on this investment. And overall, it leads to EUR 69.5 million. Cash flow from financing activities, minus EUR 45.4 million over minus EUR 22.7 million in the previous year. This increase over the previous year is based on the line item other payments with interest nature, so minus EUR 23.3 million. That would be the investments we made in the context of white spots in the first half of the year. So agreements with the BMDV where license payments were staggered while we're obliged to invest EUR 50 million. The result is a free cash flow, which is almost on the previous year's level, minus EUR 85.5 million (sic) [ minus EUR 58.5 million ] this year over minus EUR 58 million in the previous year. Now on the next page, you can see the bridge between EBITDA and the free cash flow. In the first half of the year, on the left, EUR 326.6 million are indicated. And then you can see also the receivables, EUR 51.2 million and change. The reduction of inventories, EUR 51.7 million, so this is positive. And also the contract assets being reduced EUR 43.6 million; deferred expenses, EUR 232.7 million, that's due to the payments to Deutsche Telekom in the context of the agreement we have with them; then trade accounts payable, minus EUR 45.4 million; a positive effect from other working capital, EUR 6.3 million; taxes, minus EUR 123.4 million; and last, but not least, CapEx, minus EUR 34 million. And it leads to the minus EUR 48.5 million (sic) [ minus EUR 58.5 million ] in free cash flow. Now let's move on to the outlook. As you all know, we updated the forecast at the end of last week. This year, we believe we will have service revenues of plus 3%. So we want to increase it to around EUR 3.33 billion. Our previous assumption was EUR 3.37 billion. EBITDA in the group is to be increased by 5% to EUR 686 million, and that includes EUR 14.3 million of out-of-period expenses. So far, we had calculated with EUR 720 million EBITDA. Let's look at the segments. Access, plus 9% to EUR 860 million. So far, the forecast had been around EUR 880 million. In the segment 1&1 Mobile Network, we can see higher out-of-period expenses, around EUR 14 million. And that is why the forecast for this year has been adjusted to minus EUR 174 million start-up costs. And previously, we had calculated with minus EUR 160 million. Cash CapEx for the business year, we are expecting EUR 460 million. And previously, we had expected EUR 380 million. And this is due to the fact that CapEx is now being brought to us. So we have material that will be put on stock, and it had been foreseen for 2026. So this is just a phasing effect really. Okay. And with this, we can start the Q&A session.

Nobody wants to start. Mr. [ Eisel ] just raised his hand. Please go ahead.

I was a bit preoccupied when we had that network failure, and I read about it. I think what would be important would be to get a bit of context on it. I hear some saying it's in a network area. Failures of a couple of minutes are already relevant to the statistics, and an hour or even 3 days is a complete disaster. And then I thought to myself, well, what would happen to me if my mobile network failed? It would be quite a disaster for myself, and I'd be quite angry and I would wonder how this could have happened. That's why I could imagine that the matter, we worried quite a lot of people and had a big outlash. But I think, I mean, the competitors will, of course, jump on that if one of the competitors show some weakness. So what's more important to me than any implications on the sales topic is to look at the network expansion. To what extent will we struggle with that in the future? Maybe you can outline what actually happened and to what extent you were able to understand why that happened and to what extent you can avoid this to happen again in the future. We also read that this transformation, so the migration of customers into the new network was, of course, slowed down, which is completely understandable to me. If I need to work on one thing, I, of course, cannot pull full throttle on another topic, which was migration, in this case, if I don't know what happened in the first place. And we saw this effect happening afterwards. But I was just wondering, do you have everything under control? Do you know what happened? Do you have a process to investigate this? Has this process been completed? Can you say anything on that also for the future? Because we, as investors were, of course, shocked and we said, "Well, we're already doing surgery, open heart surgery," so to say. So what can you say to that? What measures have been taken? What's the analysis? Can you disclose this analysis? Can you disclose what happened? I've heard wild stories of, "Well, they don't even know where their customers are. The system is not able to always know where the customer is located." The old story of shifting from one area to the other. It's more problematic from one network to the other. It's more problematic than we thought. So instead of listening to these rumors, it would be great to get some information from the source. Mr. Dommermuth, it would be great to get some information from you, to hear your perspective. We heard on Monday when the stock market crashed in general, there were a lot of stocks that recovered, also growth stocks and others that were impacted more heavily. Unfortunately, we got the ad hoc notification in the middle of that stock crash, and the stock decreased by 20% and hasn't fully recovered yet, if I look at the stock prices at the moment. So it would be good to get a good feeling about this. Because if everything is going according to plan, those are good stock prices, but we still need the transparency that you have everything under control. That would be my main question. It was a long question, but I think it's relevant to all of us.

All right. I'll try to answer as well as I can because I'm not a technician. I'm not an engineer myself. So the failure, of course, this should have never happened and should have never been that long. Like any failure in a mobile network, it came as a surprise to us. It was unexpected, and we analyzed it afterwards. I'll try to put it in simple terms. We have 2 data centers that are operating at the moment. Third and fourth will be operational very soon, and they're redundant, meaning when one data center fails, the other one will take over. And this redundancy didn't work out not because we don't have enough servers or the software fails or because we haven't tested it before but because in the data center that failed, we had a file that we needed in that other data center. So that was a bad configuration. We didn't have the file that we needed when this 1 data center failed. It took us a while to get this up and running again, almost 1 entire day. And that's way too long. But for the issue that we encountered, we were happy that we even managed to fix it in the first place. Of course, we learned from it. We put a lot of audits on top of this. We did some follow-up, and we assume that this will not happen again. Now you said it took 3 days, and you're right. We had some further implications for 2 more days. The reason for that is that some customers could get in and others couldn't, and that was due to the capacity that we had at this point in time. This capacity issue we had was a 2G data signaling issue. Now you might be wondering why do we even have this 2G data signaling traffic? Well, we solved the issue by shutting this off because you don't need any data on 2G. 2G is old phones, old cell phones. Nobody is trying to enter the internet through these phones. If I think about partnerships that we have now, they don't even include 2G. But we do have this 2G and the signaling of this 2G data traffic because all devices try to enter at the same time. Again, I'm not an engineer, I'm not an expert. I'm just trying to explain in my own words. This is what we struggled with for 2 days. Had we shut this off the very first day, then we would have solved it sooner. This was a learning. And we did that together with our team, the experts from Rakuten and the engineers from Mavenir because that's an issue that doesn't only apply to OpenRAN technology. Our OpenRAN technology or software wasn't even affected by this because one might say, "Well, your pioneers, you're doing something new," but no, it affects the core. And the core comes from Mavenir, and Mavenir has a lot of these cores operational and they're a reliable supplier. But it simply wasn't configured correctly. Now we configured it correctly, and that's why this issue will not come up again. I'm phrasing it like this very consciously. I'm not saying there will never be an issue anymore because we see it with other suppliers who have been around for 30 years. They still have failures. That's why there is never 100% security. But it should happen rarely and it should be fixed quickly.

And you said then you had issues with your migration of existing customers.

Yes, that's correct. Because we needed to make sure that this network could fully operate in all dimensions and that we won't realize tomorrow maybe, oh, there is another issue in another place, wrong configuration in another place. So we put a couple of audits on this. And we found a couple of minor issues that we then fixed. I mean the network is operational the whole time. After we fix the failure, it was operational again. But we said, for security reasons, we are not going to add on more customers on this network than necessary. New customers are necessary, but I don't have to migrate existing customers right away. That's why we said we'll leave the existing customers where they are right now, and we try to be 100% on the safe side first. Now you might say, "Oh, now you're 100% safe, you did audits, everything is running. Why are you not doing it now then?" We're not doing it now because our 2 core data centers that are upcoming, #3 and 4 have a bit of a delay. They're all set. We have servers, connections to Telefonica. The networks or the data centers are interconnected. But now we're adding the software step by step. We're doing tests step by step. So we're being a lot more careful than we were the first time around. That's why we're taking a little bit more time to get that up and running. In our existing data centers, we have a certain capacity. And I don't want to run out of time by filling this capacity with existing customers. I want to add new customers into that first, so that in case everything will take a little bit longer, I'm more relaxed, even if it takes 1, 2, 3 months more. I always want to have enough capacity in our existing data centers. That's simply precaution. We could continue migrating customers, but then we would have to be sure that everything will be finished by September because we're migrating 50,000 customers per day. And if we also add new customers on top of that, we'll fill up our capacities really quickly. That's the reason why we've been pausing existing customer migration. So in a nutshell, it should not have happened. It should have been fixed quicker. And there shouldn't have been any delays with data centers 3 and 4. But now we just want to be on the safe side, and that's why we'll just take our time.

Maybe a follow-up question in a different area. Contract with Vodafone, there haven't been any news on that. I know that you said 3 months ago that there is a binding contract or binding agreement and you're now phrasing it. So you're going through folders that need to be processed. Now it's 3 months later. And now the question comes up, it was supposed to start or it should come into 4th in July, maybe even October. Now first part of my question is, is everything on track. When will you start? Have you started? Will it happen in August, September, October or maybe a bit later? And the agreement itself. Are there any aspects that we should know about? Maybe any aspects becoming more complicated because that could always happen. When you hash out the details, one party or the other brings up a topic that needs to be escalated, just like it always is with more complex matters. As far as I'm aware, you also wanted to be very cautious with this agreement, you did not have the same issues that we had with the Spaniards, so to be very, very careful with all the details instead of having to bring things up in court later on. Maybe you can give us an idea of where we're at and what your perspective is because I hear it over and over again, it's such an abstract topic where everyone can bulls*** and say, "Well, they haven't even signed an agreement. Mr. Dommermuth always says everything is binding, but we don't get any reliable information." So today, it's a good opportunity to get some more reliable information.

We informed everyone that we have a binding precontract with Vodafone and we informed about that in August of last year. But this contract only consists of a couple of pages. But these couple of pages are binding. So it already has some information that is binding. But now this needs to be put into an actual agreement or contract, which is very complex because there are many annexes, technical annexes to this agreement, how pricing is done, for example, so it is highly complex. But I don't want to hide behind the fact that it's complex. We fully agree with Vodafone. I am not aware of any aspect where we have different opinions. The teams are now almost finished with finalizing the annexes. I think I received 2 more annexes yesterday that are ready to be read through. So I think it's a matter of days, I would say, but it's not just under my control. It's also under Vodafone's control, so maybe a matter of weeks. At the same time, all the technology is set up. Vodafone has built everything. We built everything on our side. Concurrently, we're doing a friendly user test, and we would be ready to go in a couple of days. So we said it last year, we want to start in the summer, and we will start in the summer. Now you said, well, there are a lot of details to be hashed out. That's true because we're talking about multiple billions of euros. It's about an agreement that runs for 5 years but can be extended beyond that. And so you, of course, need to anticipate what can happen along the way. Just recently, we had a CEO meeting, let's call it that, with Vodafone, where I participated and Vodafone said, "Well, you're talking about topics that we've never talked about with anyone else. Those are things that don't even exist yet. And you want to put all of that in writing now." And I said, "Yes, that's exactly what I want. I need to put in writing what doesn't even exist yet." Because if we think that it might come, if we predict this to exist in the future, then we need to be able to describe it today. But this is an issue, of course, because it's hard for the teams to make progress because it's also some abstract but more general topics, how do you describe new technology that doesn't even exist yet and that you don't know how it will be implemented in the future? How do you describe what should happen when, for example, network providers cannot provide this technology right away if there are further implementation costs, et cetera? How do we handle such abstract matters? You don't have to define it. You can just say, "Well, we'll define it or negotiate it later. We'll do it in good faith." But I don't want that. We're talking about a lot of money. And the reason why we're switching to Vodafone is not because we'll get better conditions but because I want to have less issues. That's why I want to have everything covered as far as possible so that, in a couple of years' time, when there may be different people in the company want to discuss what we meant with what we put in the contract. So yes, a lot of details, but I don't mind that because we wanted to start in the summer, and we will start in the summer. It won't improve our position. I can say, "Well, if we negotiate for another year, we'll improve by another couple of percent." Now our conditions haven't changed, but the advantage is it's also not become more expensive. The conditions simply haven't changed at all. So these details are a bit cumbersome, but they don't prevent us from reaching this agreement. The cooperation will happen just like we agreed last year.

Next question from [ Camilo of Zeus ].

Sorry for the question in English. Obviously, we're sitting here past the 1-year anniversary of the Vodafone agreement announcement with the share price almost at the same levels, which creates, obviously, a lot of frustration given all the optionality you have on this asset and the value this asset has as 1&1. So one of the two key aspects that led to this is, obviously, the lack of visibility in terms of the financial trajectory on the CapEx and the lack of communication. So maybe if you can help us understand the two aspects. The first one is that Vodafone has been excellent in sharing infrastructure with other operators in other markets in Europe to reduce their CapEx needs and increase the return on capital employed. It sounds like in Germany, it could make sense to mutualize infrastructure and invest in a smarter way to optimize footprint, reduce overlap, reduce CapEx and increase return on capital employed. Is that something you could potentially consider to do in Germany with Vodafone as Vodafone has done in other markets? My first question. My second question, still on CapEx, which I think is the key element of your story. You've increased by EUR 80 million your CapEx for this year. In the ad-hoc release, you didn't give us much visibility on what it meant in terms of phasing. So could you please clarify that this EUR 80 million extra CapEx is CapEx taken off from 2025 and 2026? And could maybe Mr. Huhn confirm that the CapEx for 2025 and 2026 should probably be a number much closer to EUR 300 million versus the EUR 360 million to EUR 380 million that the consensus has currently? And the last element, which would be helpful to get visibility on, is on the low-band spectrum lease discussions, on where you stand and by which time line you aim to provide the market with more visibility. Obviously, it's a pity the webcast cannot ask questions, but I imagine these points are very important for the market to realize the value in your company, which is very high.

Thank you for the question. Okay. You started with the stock price. Yes, of course, we're not happy with this. But of course, we are not determining the stock price. We have to make sure that revenues and profits are right, and we have to make sure that we don't have any failures, of course, as we have just discussed. But of course, we would wish to have a better share price here. So how can we contribute to this? You have just asked, does it make sense to share infrastructure with Vodafone, so network parts, constructing them only once? Yes, from a CapEx point of view, it does make a lot of sense, particularly in rural areas because they are also -- sometimes the network usage is not that high. I know that we do have that in other European countries, yes, but we haven't entered into any discussions yet on this. So I'm not going to tell you that we will never have those discussions. Of course, it also depends on how Vodafone sees the world. But so far, we haven't had any discussions on this on the sharing. Okay. Afterwards, you asked what about the EUR 80 million additional CapEx, is that phasing? So let me try to explain it. We have a partnership with Rakuten and we have defined Rakuten as being our system integrator. So they also have hardware on stock for us, so antenna or service, for example. At the beginning of the partnership, we set up a rollout plan, but of course, it's not a secret that it has failed, this plan. We would have wanted to be further than we are with the antennas. It is now starting to go well, but there's a delay, of course. And Rakuten, they needed to upgrade their stocks. And then the question was, how are we going to deal with this? From a contractual point of view, we are only obliged to purchase goods or purchase hardware once it is integrated in the construction. But if you want to have a long-term partnership that works well, I do understand also if a partner tells me, "Well, I have had this on stock for quite a while. So why don't you just take over part of it," meaning those elements, which you will need over the next couple of months. So that's what's behind it. It's the pure phasing effect. The antennas, the servers, they're already there. They are in Germany. But the storage is being paid today by Rakuten. And tomorrow, we will be paying it. So what's the advantage? The equipment is already in our ownership. We have better access. But of course, I don't want to sell it as a huge advantage. Of course, it would also work out really well if it was in Rakuten's ownership. So it wasn't our desire to really take it into our ownership. But I do understand it and find it plausible if someone tells me, "Hey, there's a great delay. We have been financing it for a long time. Maybe we can discuss the change in ownership." So I do understand it. And I think it is also part of a good partnership to say, "Yes, you're right. We should have really purchased it earlier." And then we will do it. However, it's not additional hardware. So we don't need more antennas than originally planned. And also the price per piece has not increased. It is just a phasing effect. And maybe Mr. Huhn can also say something about the CapEx in 2025 and 2026. And then you also asked about the low band and the status of the discussion and more visibility. Well, all other network operators, and also we, have answered the consultation of the Federal Network Agency. We submitted our assessment. And usually, it is being published on the website of the Federal Network Agency. However, it takes a couple of weeks for all of the submissions to be published. I think it hasn't happened yet, but it won't take too long, and then they will be visible because the Federal Network Agency will have to make a decision as well in the next couple of months, and that's why they have to publish the submissions. So in a nutshell, what have we uploaded? We are stating that we are generally open minded towards the proposal of the Federal Network Agency to extend frequencies and to make sure that 1&1 can do a sub-usage because that is the proposal of the agency, that's generally acceptable. However, the question is what will the terms and conditions be? And what is the frequency amount we would get in order to serve all our customers. So proposals have been made by the agency, which we believe do not go far enough. Well, but that's the process of discussions. Whether it's telecom, Vodafone, Telefonica, if they said, "Well, the proposals go way too far." Well, then we will have to see where the procedure will end. It will be up to the agency after all, not up to us. But generally speaking, we are open-minded. We think that the direction is already a right one. Last year, in the first proposal, we weren't included. So now we will be taken into account. But to what extent, that needs to be discussed still. So of course, we don't want to create any blockades here. We need frequencies in order to make sure that our customers receive proper and high-quality services. So we don't really care whether we purchase the frequencies or whether we lease them because that is also being done. The 2.6 frequencies are leased with Telefonica, and it is just as good or bad as if they belong to us.

Okay. And I will answer the question on the CapEx. I can confirm what Mr. Dommermuth has just said, the EUR 80 million are a pure phasing effect. For the CapEx, let us distinguish into CapEx development without passive infrastructure. So for passive infrastructure, it is our approach that we want to find a partner to whom we will sell the hardware and do leasing agreements so that we don't have the CapEx. We haven't found a partner yet. Now looking at the CapEx Rakuten active infrastructure. In 2025, we expect to have lower CapEx than in 2024, so around 10% less. This might seem to be fairly little at first glance. However, in the summer, there will be a shift in the CapEx because over the past years, we invested mainly in data centers and servers, and those investments will go down. We don't foresee a lot of them for next year. However, the number of antennas and sites we are acquiring and commissioning, we have seen that. And that is why the investments for active infrastructure will go up. But all of this is contained in the figure. So despite the increased number of sites that will be commissioned for Rakuten active infrastructure, we believe that CapEx will be going down slightly. If we did not manage to find a partner for the passive infrastructure via whom we can avoid to have this additional CapEx, so selling the sites, this would mean that we might require around EUR 100 million, EUR 150 million CapEx more next year, depending on the amount of sites we can get constructed. But it is still our objective to find a partner so that we can make sure that just as the colo sites via Vantage and APC can be leased rather than having them in our ownership or property.

And if I may add to this, for the sale and leaseback of radio towers, as Mr. Huhn has said, that is something which is quite a positive business, and that's what we are targeting. They don't need to be in our ownership. But the question is always, what are the terms? Currently, we are obtaining quite cheap financing. Interest have increased overall, yes. And of course, that is something that needs to be factored in today as well. So the expected leases or lease increases, they factor in increased interest. And on the other hand, we can say, well, financing conditions for us are quite favorable. So we don't really have to sell our radio towers immediately. So it might happen that we sell them if the terms and conditions are good for us, but maybe we will also keep them. So it is not our objective, but also it wouldn't hurt us because it needs to be worth doing it in the end. And it might they might seem to be small amounts at the beginning, let's say, EUR 3,000 per year per radio tower. But if you have 12,000 of it, it makes EUR 36 million. And if you do it over 20 years, you arrived at, well, EUR 720 million. And then, let's say, TowerCo increases prices because they act as if they were a letting agency. And they want to have compensation for increased prices every year. So for the ground trend. I do understand it because after all, the owner also needs to receive the compensation for the price. But I don't understand it for the infrastructure because it was purchased once and it has been paid in full. So it doesn't gain in value. However, that's the business model. Just as if you were to build a house, you expect to receive more rent even though you have already paid the price for the house. So the higher you set the interest rate assumptions and also the inflation rate, the higher it diverges really. And that is something you need to take into account. So we are now in a comfortable situation to take a free decision here. We are not under pressure. We don't have any debt at 1&1. We still have cash with United Internet. So it would be difficult to explain why we entered into a nonoptimal deal just in order to avoid CapEx.

Okay. Thank you very much for your questions. I think they were quite comprehensive. And I think we didn't have any more comprehensive Q&A at our Annual Accounts Press Conference. So if you still have any open questions, please feel free to address them to us after the call of United Internet, which we will have after a short break. So thank you very much, and all the best to you. [Statements in English on this transcript were spoken by an interpreter present on the live call.]