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

Good afternoon, ladies and gentlemen. Let me welcome very warmly to our conference taking place for the first time in person again on behalf of our CEO, Ralph Dommermuth; and our CFO, Markus Huhn. As usual, you're being taken through the presentation. And after that, Board members will explain the strategy to you and will explain the forecast. After that, we'll have a Q&A session as usual. Let me be very brief and wish you an interesting session. And I'd like to give the floor to Mr. Dommermuth.

Well, good afternoon, ladies and gentlemen. Thank you very much, Mr. Kyle. Welcome to our figures for H1 2023. I would like to first introduce to you the company development and the status of our mobile phone network, and then Mr. Huhn will give you the figures -- financial figures for the first half of the year and the outlook for the second half. Let me start with the product portfolio. You know our products were strong in our -- in the DSL business with VDSL and ADSL and fiber to the home, and we do that together with our sister company, Versatel, and we take the -- used the last mile by Deutsche Telekom or city carriers. We are the fourth largest mobile phone provider in Germany with just under 12 million customers. We have a broad market coverage with different brands, our main brand, 1&1, then we have the core brands, GMX and WEB.DE and our discount brands. We're currently in the process of being a 5G network, and I'll get back to that later on. In the first half of the year, we were able to conclude an extra 180,000 customer contracts. We have a total of 15.96 million customers and an increase of 230,000 mobile Internet contracts and a decrease -- slight decrease in broadband lines. We've witnessed this decrease for a number of quarters. We expect that we can stop this decline in the fourth quarter now. That's due to the fact that our old ADSL customers, where we have a high cancellation pressure that we can migrate them to VDSL, and those are contracts that we can handle very well. And that's why we expect contract numbers to stabilize in quarter 4, and we won't have a further decline here anymore. We were able to increase our revenue by 2.1% in the first half of the year by a small margin to EUR 1.93 billion. And we have a service revenue and other revenue of 10%, but that's a low-margin business that we have there. The EBITDA in our core business, in our access segment increased by 2.2% to EUR 392.8 million, that's within expectations. Maybe a few words on revenues. That's also as expected, I forgot to say that. So we see an increase of revenue in 4% in the second half of the year so that we will be in excess of the service growth of 2.2%. Now as I said, the EBITDA increase of 2.2% is within our expectations. In the mobile network, where we have our own, we're building our mobile network, has decreased by EUR 40.8 million compared to last year. CapEx also, due to the buildup of our network, increased to EUR 81 million, EUR 6.2 million of this in the Access segment, it's a small percentage, of course, and EUR 74.8 million in the mobile phone network. Now where do we stand in terms of our mobile phone network. You know that we're building a virtualized OpenRAN for the first time in Europe. So we don't use any Chinese products. We only use products by Western producers or from the Far East. We've been active in this area since the 28th of December of last year. So we have fixed wireless access instead of land lines. We have 2 core data centers of 4 data centers, 22 of 24 decentralized edge data centers are in place. 74 of approximately 550 regional far edge data centers. So that's our first milestone. And we have taken over 193 antenna sites. Taking over means that it's ready for installation, RFID. And 40 of them are already in operation, i.e., they've been connected to the fiber optic network. In the first half of the year, we connected our network with all national -- international networks relevant for us. We are in test operation with a national roaming with Telefonica, will be made available by Telefonica. We have a friendly user test and operation with over 300 external customers. And by the way, we -- we're also successful in court against Deutsche Telekom. We can still say that we are building the most advanced 5G network in Europe. Now the cornerstone data again for our calculation of the network. With the network, we want to compensate for a current wholesale expenditure, and that means that the network should cost us much around about as we are now transferring to O2, our partner for the infrastructure. So the software, the data centers, edge data centers, et cetera, will save all that cost by no longer having to pay by the minute for voice, but that we produce it as voice traffic. That's particularly low cost because voice is not very data-intensive. We also save money in international roaming. We have upfront costs now that are higher than the cost that we can achieve as network operators. We can confirm this today because we can see that this will actually pan out the way we had anticipated. The operation cost of antenna sites are to be covered by savings on national roaming antenna sites. We have to depreciate for the antenna for the mesh for the rental. And there's power requirements, et cetera. So antenna has to produce a certain level of gigawatts -- gigabytes, sorry. It has to produce as many gigabytes multiplied by the purchasing price as the antenna costs on a monthly basis. And as far as we can tell today, this will work out. We bid on frequencies, and we were successful. For frequencies worth EUR 1 billion, we will have additional locations that we didn't have available as operators -- network operator in the past. So that is all according to plan. We will take over another additional 300 co-location sites this quarter and we'll make another 200 leases for new locations. You can see it on the graph on the right-hand side. In 2022, we took over 55 sites in the whole year -- quarter 4 of '22. In the first quarter this year, we took over 95 million, and then we added another EUR 193 million. Now we're adding another 300, so we'll have a total of 500 by the end of the year. The light blue area shows the lease agreements of new locations. We are taking over new sites that need to be built from scratch because we can't use existing co-location opportunities because either no co-location is available or they're not suitable or the used capacity. So we've taken over -- we've signed lease agreements for 98 sites in the second quarter. And in the third quarter, we will sign another 300 agreements. We'll have a total number of 1,000 sites by the end of the year. We had anticipated 1,200 sites by the end of the year, but we had a cancellation of delivery of about 300 sites by our largest expansion partner in June, and therefore, we won't be able to reach the target of 1,200. We can compensate for part of this loss, but we'll only wind up at about 1,000 sites. And I've also put down here that the ramp-up is underway despite further delivery failures. We will have 500 new sites, 300 co-location areas, 300 leases that will be managed by us. Then the development of antennas, the fiber optic cable by Versatel and also with new leases, we can build new masks before we sign a lease contract. Of course, we check that the location is appropriate in terms of the radiation levels that we have to prove and the statics of the roof, et cetera. So we no longer rely on lease contracts. That's a critical point. Now we only need to install active infrastructure. And that is the beginning of the funnel of course. And it takes a long time before the fiber optic cable is in place until the antenna has been installed. But if we add another 500 sites every quarter in a funnel, then of course, every quarter, we will have also 500 sites coming out at the other end of the funnel every quarter. So if you look at the numbers here, you have 500 per quarter is 2,000 a year or so we will -- we are just -- we are on track to reach our target. We want to achieve 3,000 next year. And it's not illusory. As you can see, we're already right on track. And so I think this is a very important slide indicating that we are gaining traction here and that we are really making progress. We reported yesterday that we made a new contract with Vodafone. It is a binding precontract. Some sites need to be defined before certain specifications are laid down. So standards, terms of duration, cancellation periods, et cetera, that's all put down in this contract. So Vodafone will give us not just -- every access to their network. It starts in July 2024 by the earliest, but no later than the first of October 2024. It runs for 5 years. Extension option for 2 times 5 years and then a 3-year transition period on top of that. So we're talking about a capacity model, not a price per gigabyte, but percentages of the Vodafone network. So of course, that means a certain number of gigabytes, but that changes all the time because it is being extended. The network of Vodafone is being extended [ at the end ]. So if the network has a capacity of 100,000 gigabytes today, then would be 1% equivalent to 1,000 gigabytes. So if it has a growth rate of 30%, it would have -- it would mean 1,300% of gigabyte next year and so on. And that's the interesting thing. We keep paying the same price. So that's -- we will have access to a virtual network as it were with the benefit of having all the technologies available, with the benefits of network growth, whatever that means for the coming years, and new technologies. But on the flip side, of course, and that is written in the next sentence, with a price increase or decrease in tandem with Vodafone -- the Vodafone cost base. So every few years, it will be analyzed what the cost of the Vodafone network has developed like. So if it has become 10% more expensive, then we have to start paying 10% more, that's fair. If the cost decreased by 5%, then we will pay 5% less. And that is no more than fair. So we will accompany the bolt-on costs, 1:1 for our 1% share of their network. I was asked by the press earlier, are you saving a lot of money that way? And I have to say, no, we don't because the costs that we pay Vodafone are about the same that we pay to Telefonica. If you look at the next 5 years. But it's a much better contract because there are no hidden costs here that could surprise us in midstream. There's no limitations. And the discussion of what is the right price per gigabyte that we sometimes had to take to arbitration and with -- where we had to get expert opinions over year-long processes that took us a lot of effort, high legal costs, et cetera, that will no longer be given. So we have a lot of security here for the next 19 years. So 1 year before we start, then 5 years to start with 2 times extra 5 years and the 3-year transition period. So over that period, we can provide appropriate services to our customers while we keep developing our own network. I said earlier that this will take until July to October of next year before it starts. Now what are we going to do in the meantime? We want to start our Internet service in the September of this year. First of all, we would like to cater to our 4G customers and all the discount brands, they're all running on 4G one-on-one customers. And old contracts also have 4G. We will produce 4G with our own network and with the national roaming with O2 Telefonica. We will compensate for the lack of coverage that we have. 5G until provision of Vodafone National roaming next summer, we will solve with an MVNO model. And once we have national roaming by the summer of next year, then, of course, if we have 5G national roaming, then, of course, we'll make the network for 5G networks or services available and make that available to new customers. That's the plan for the transition period. Now that's due to the fact that Vodafone takes about a year in order to build up national roaming, that was the same time that Telefonica took to install that, so that makes sense. The condition for this is that the Federal Network Agency gives its approval. They approved parallel operation by the end of this year. But if we want to be MVNO and network operator until the summer of next year, we need an approval until the summer of next year. We applied for this approval yesterday. This application is based on the conditions for frequency allocations which says that there can be parallel operation for a transitional period if the technologies are different technologies, and we're separating it here. 4G with our own network and 5G with our future network, and we'll have to see how -- with Vodafone network, so we'll have to see what the Federal Network Agency will decide in this context. But we believe that we can argue our point very well. Next steps, as planned in the next weeks and months, we'll start the third and fourth core data center, then we will take over national roaming from Vodafone. We'll start with the smart fee structures. And in quarter 4, we'll start with the start of migration of existing customers. We will migrate 5G customers later, as I said earlier, but our agreement with Vodafone allows that explicitly that we approach it this way. Okay. Thank you very much, and I can then ask Mr. Huhn to present the financial figures.

Good evening, good afternoon, ladies and gentlemen, on my behalf as well. I would like to start with P&O overview, we see the increase of revenue of 2.1% and Mr. Dommermuth had commented on that. The sales of -- cost of sales, we see an increase to EUR 1,388 million, an increase of 4.8%. In this position, we have start-up costs for our 1&1 mobile network. That was EUR 12 million beginning last year and the first half of this year, EUR 54.6 million. And that leaves us with a gross profit of the turnover of EUR 604.4 million coming from EUR 627.3 million before. If we take out the start-up costs of the net from this, the gross result would have changed from EUR 639 million to a plus of 3.1%. So compared to the turnover, we would have grown overproportionately. The sales costs have increased from EUR 248 million to EUR 245 million (sic) [ EUR 254 million]. And the administration costs have gone to EUR 57 million. That has depreciations of EUR 8.1 million and EUR 4.2 million coming from the year before. If we eliminate this position, we see the administration costs without the 1&1 mobile network has a slight decrease. The other operation activities have increased to EUR 13.5 million. That is mainly for fee collection processes, an increase of losses on receivables have increased to EUR 52.4 million, which is because of the hardware [ turn ], value adjusted by share to the expected loss in the balance sheet. As in the first year, the increase in hardware was increased by 10% roughly, that has increased by 6.4% as well -- 6.5%. The result was EUR 254.1 million coming from EUR 287.7 million in the year before. That means a slight drop due to the start-up cost of the 1&1 mobile network. If we eliminate these, the result of the activity would have come from EUR 299.7 million to EUR 300 million, which is an increase of about 3%. The financial result was at EUR 3.4 million in the first half year after a minus of EUR 3.2 million in the first half of '22. The positive development is due to the element of interest. We have about EUR 8.1 million interest profit due to the financial management at [ United ]. The profit before taxes is therefore at EUR 257.8 million (sic) [ EUR 257.6 million ]. The taxes are EUR 77.6 million and the consolidated result is EUR 180 million. Let's look at the balance sheet. The sum has increased from EUR 7,257 million to EUR 7,502 million. And in the short-term assets, we have a slight increase of EUR 36 million from EUR 1.850,5 million to EUR 1.887,1 million and the long-term assets due to the contingent payment to Deutsche Telekom and on payments for equipment of our mobile network equipment have moved from EUR 1.5 million to EUR 1.6 million. On the liability side, we see the short-term liabilities increasing. The long-term liabilities were above the level of the year before with EUR 1.130,2 million and the equity capital due to the positive result in the first half has moved from EUR 5.579,8 million to EUR 5.751,8 million. The cash flow, the net payment of the operating activities were at EUR 23 million in the first quarter compared to EUR 240 million in the year before. The cash flow from investment activities was at minus EUR 0.8 million coming from EUR 219 million in the year before. Cash flow from funding activities is on the level before with a minus of EUR 22.7 million compared to EUR 19 million in the year before. And the free cash flow at minus EUR 58 million compared to EUR 189.1 million in the year before. The main difference, which leads to that first half year comparison of '22 and '23 in the first and the last position is due to the fact that the payments from the contingent contract in this year was already made in the second quarter this year. And in the year before, in the first half of the year 2022, this payment hasn't been made. And last year, we only did that in the third quarter. And if we take that effect out, then the net payment in the operations, we would be a little over EUR 240 million just like the cash flow from investment would have been at the level of the year before, and the free cash flow would have been slightly better than the EUR 189 million that we've had in the year before. So at the end of the third quarter, if we look at the cash flow then, you're going to see that here, in all positions, we are roughly on the level that we were in the year before. Now we have the bridge of the EBITDA to the free cash flow. You see the EUR 352 million EBITDA in the first half of '23. Negative balance EUR 42 million from liabilities and assets. The occurred expenditures due to the contingent payment of the Telekom, mainly with minus EUR 254 million. The balance from liabilities of associated companies is positive with EUR 61.8 million. The assets in the minus EUR 3.2 million. The other working capital has an effect of minus EUR 23 million. And from that, we have taxes of EUR 68 million and CapEx with a minus of EUR 81 million leaving us at a minus of EUR 58 million free cash flow as of the end of the first half of 2023. Let me give you a brief forecast for the financial year 2023. We confirm the forecast so far the service revenue with a plus of about 2% to about EUR 3.23 billion. The EBITDA with about EUR 655 million with a plus of 4% in the access segment to about EUR 775 million. The segment 1&1 mobile network with a start-up cost of about EUR 120 million. Our customer contracts where we expect a growth of about 500 new contracts for this financial year and a cash CapEx mainly from the building of the network with about EUR 320 million for this fiscal year. Okay. Thank you for your attention so far. We are happy to answer your questions from now on.

We'll start with [ Mussato ] of Societe Generale on the left side, please.

Let me start with a couple of questions and a clarification, if you don't mind. First of all, I'd like to know what are the costs involved in the cancellation of the O2 contract. O2 Telefonica Germany made a press release yesterday pointing out that there were contractual obligations, and I'd be interested in these costs very much. The second question is, well, one could argue that with this national roaming agreement with Vodafone, the need for the low-band range is very much reduced, so access to -- because you have access to all technologies on the Vodafone network now, could you elaborate on that a bit? The license agreement, this 50% coverage, could that be covered that way and have the rest with Vodafone, thus gaining access to the low band range this way, and therefore, be less dependent on an auction? And the clarification is concerning economics. It says savings for national roaming. I'd like to know is that savings compared to the MVNO or is it savings compared to the national roaming with Telefónica Deutschland? What's the baseline for this savings?

Well, let me start with the last question. The costs in the MBA agreement are identical with those in the national roaming with Telefonica. That means you can interchange the tax, the antenna saves national roaming in its range and also sales MVNO capacities, that's identical because the process are identical with Telefonica. And as I said, the Vodafone prices don't defer that much. That was not our negotiation objective to optimize this way because we can handle today's pricing structure quite well. Yes. Of course, you could use different yardsticks now. If I look at how we purchase traffic, how we purchase international roaming in the European countries, we pay about 44% of what we pay for roaming in Germany. So it's a very lucrative business for the network operator making roaming available. But with our cost structure, we can handle this very well with conditions that we have today and the conditions that we will have going forward. It all fits our business plans. But there's no, yahoo, Vodafone is much cheaper, no. It was mostly about the contractual agreements, not so much about the conditions of the financial pricing. Then you ask what are the obligations resulting from the cancellation of the O2 agreement. There aren't any. We have an agreement with Telefonica, but no obligation to have -- purchase certain volumes. We can increase our volume within certain ranges, and we can reduce them by 100% by the [ fifth of 25]. After that, we have a 3-year transition period. And there is no minimum volume during this period or pricing structure there. So this does not involve any cost -- additional cost for us. That was not my reading of Telefonica's press conference yesterday that we still owe money to them. No, they said that the contract will continue until 2025. And in that period, they will still earn money with us, and that is true because we're only transitioning our customers from Telefonica to Vodafone by the by. So we will have business with Telefonica up until 2026 and even possibly beyond that. We will buy services from them and pay for them, of course, but there are no -- as contractually agreed, but there will be no compensation payments or penalties or whatever you want to call them. And you also asked, doesn't the need for low band is not reduced. And we don't only need that for distribution in the field, but also for in-house distribution, you couldn't use your phone in this room with the frequencies that we acquired. These frequencies have characteristics that stopped their propagation just beyond a window. That's why we need low band for indoor operation because most traffic happens, of course, indoors because you don't normally use your mobile access while you walk out doors, but it happens in houses, in buildings, in subways. And so yes, the need for low band frequencies is still there, and we need to acquire them. You might say you can compensate for that, simply continue using indoors roaming, but the antenna always cost the same amount of money. The glass fiber connection, the rental for the site, the antenna itself, that always costs money, the data center, the core system, the software, all cost the same amount of money. And our antennas are all equipped such that we can handle all frequencies with the exception of low band. And I think 40% to 50% of low band -- 50% or 40% of our antennas are equipped such that they can handle low band. You don't normally equip every antenna location with low band because our distances between our antennas is very small. So we need to equip about 40% to 50% of antennas with low-band capability to have full coverage, and that's what we anticipate immediately. We only have one additional position, an additional item here. That's the acquisition of the frequencies. And if you add up the rental depreciation, et cetera, so the antenna may cost EUR 1,000 a month. That's just a figure plucked from thin air. But then it's, of course, bad if this low-band frequency of this antenna only harvest EUR 500 worth of traffic because you don't have the low-band access. So that way, we can make it profitable because low band means that without low band, antennas are only partially profitable. We could compensate for that with low band, but it's a calculatory thing. So we have to calculate how much do we pay for roaming deep indoor, and then you have to depreciate for it. And then we know we shouldn't pay much more for frequency band during our auction because it wouldn't make sense. If we can get it cheaper then it makes sense economically. But our objective, of course, is to have our own frequencies because the unique selling proposition of our network is only possible if we can achieve that with our own frequencies. So right now, customers can use their phone right by the window, but if they sit down on the sofa, they can't use it anymore because then they use the abilities provided by the Telefonica or Vodafone network. And that's not our objective. Our objective is to create a new user experience, which is different from the other networks so that customers cannot use their phone not only just by the window, but in the entire room. That's what we need the low-band frequencies for.

Martin Hammerschmidt from Citibank, please.

I'd have two questions as well. One is on the calculation basis of the network cost for Vodafone. I think most of it is operational expenses, but also network building, 6G upgrade, spectrum, if parts have to be replaced, how are you going to be involved in these costs? And the second question is the price adjustment of Vodafone. You've mentioned that's going to happen every third year. Just for explanatory reason, was that an example? Or is that a real figure? And so does that mean you have a fixed price percentage for the next 3 years and only the network cost will drive the prices?

Thank you for the question. I can't give you the details of the contract now because the contract doesn't allow me to share the details here. But I give you the 3-year example. I can't confirm that it is 3 years, but expected not to be completely off. And from the point in time where we have a price increase, there is a percentage -- a fixed percentage of the price until the next adjustment. Let me take 4 years as an example, and then we'll see again how did the network costs develop compared to 4 years ago, and then that is added for the next years. So we have a follow-up effect here. During the phases, we have no price increase, but after it. And then you ask, what does this include? And you listed a number of positions. And I would say it's all there, except from 1 point. And if you think what you just said, 1 is obviously not in. I don't want to confirm, but assume the spirit of the contract is it includes everything that we need for standard business. So that would mean -- they've new towers are built up, equipment is replaced, then 1&1 would be involved in that cost. If new towers are built up, then the operation cost was increased, that means that we have higher depreciation. If depreciated equipment is replaced and that is depreciated. Again, the new equipment, if the energy cost increased, the frequency cost increase as well. But also if the frequency costs drop or the operation coming down with less staff or a depreciated component is replaced by a new component which costs less and that creates less depreciation, that is in our favor then.

We'll carry on with Steve Malcom.

Sorry, apologies for my lack of German. I hope the English translation is accurate. A couple of questions. Just following on from Martin's question. Just on those costs and what you're seeing at Dommermuth. Do we assume that if spectrum costs are -- rise or expensive that, that will be reflected in your costs with Vodafone. So clearly, with the upcoming auction, that ends up being expensive. You would end up paying more on the roaming side. Is that one of the things that we should think about in that calculation? And then secondly, just on the sort of SIM swap, as you migrate the customers from the Telefonica network, the Vodafone network. Can you maybe just walk us through the sort of the likely steps over the next couple of years? I mean, clearly, you have some obligations anyway at your network to move customers off the [ NBM ] to the roaming agreement. And your initial thoughts on how that's going to play out and by which point do you think first all your customers will be on the Vodafone network, that would be very helpful.

The spectrum costs are included, and that means if spectrum prices increase, that will concern us. If it is cheaper, we will participate as well, national roaming from Vodafone. And the migration from the MVNO contract. We do assume that with the Vodafone or the phase over to Vodafone, we will start next summer. If national roaming is provided by Vodafone. And at the same time, we can then look at roaming traffic that we already have, moved that to Vodafone. We don't need any customer participation in that. And the change process as well does not require any activity by the customer because we have so-called dual profile SIM cards. And that means a card has two profiles and over-the-air, we can transfer a second profile. And when the customer switches on their phone, the phone won't use the first profile but the new one. And then we will -- if the first one is the Telefonica MVNO profile and the second is our network, we can switch over without the customer doing any activity on their phone. And we can do that today with Telefonica SIM cards, but we can't do it with Vodafone SIM cards as of yet, but the contract has an agreement that we will provide -- be provided 5G MVNO cards and that will be dual profile SIM cards as well. So that customers that we get in the Vodafone MVNO, they can switch over without any activity required by the customer through our network.

I mean, what proportion of the base have dual profile SIMs today? And at what point do you expect all of the customers to have dual profile SIMs?

I would say today, about 90% of the customers have this dual profile SIMs and prospectively, it's growing because we only provide these. And I don't worry about migration processes of customers who don't have that. We are very trained in that with just hundreds of thousands of customers from our old E-Plus MVNO contracts, and they didn't have a dual profile SIM. And we move them simply by sending them a new card, putting in the new card in their phone in the right moment and importing the phone numbers and so on. Our team did a great job there. We didn't lose -- we hardly lose -- lost any customers. So nearly all customers will be provided with dual SIM cards and those who don't have them will be left till the end, hopefully, that will be fewer then, but we have done this exercise just a couple of hundred thousand times without big issues, so we expect that for the future as well.

It's Adam Fox-Rumley from HSBC. I had a couple of questions, please. Firstly, I wondered if I could ask you to talk about the processes of moving from the leases signed sites, so the kind of light blue boxes to the dark blue boxes in your chart. I'm sorry if you went through that in the presentation, but I'd love to go through that again. And what does the regulator think or do you have a view on what the regulator thinks about announcing leased sites given that you do have contractual obligations to roll out a certain number of sites. And then the second question I had was, once again, you've mentioned that Vantage your largest tower supplier hasn't delivered the number of sites -- that you're not expecting them to deliver the number of sites this year that they promised you. They appear to continue to let you down. You're in dispute with them. Is there any scope at this stage for compensation? Or can you talk a little bit more about how that relationship is progressing because I think previously, you were quite confident that increased oversight would really deliver the pace of tower build-out that you had previously committed to.

My understanding of the first question is that to what extent can sites be counted under the obligations of the specifications? There is no distinction whether we build our own site or whether we lease it. So the sites that we showed on the slides are composed of leased and owner-occupied sites. And both of them are added up to show whether we meet the obligations from the auction.

Is there a distinction between sites that are kind of delivered to you and ones that are live though? Because there's a big gap between the numbers.

Yes.

What is that?

Well, of course, only the live sites are counted. I think we have to say that quite clearly. We are behind schedule with our extension plans. We wanted to have 1,000 sites up and running live last year and our agreements with our suppliers would have allowed for that. But we failed to achieve that target, and we won't be able to achieve this by the end of this year. We will have 1,000 sites ready for installation by the end of the year and a few hundred of them will be fully operational. But there will be a large number as well where we still are waiting for the fiber optic connection or the antennas. Let me give you some explanation of the background here. If or when we develop our own site, then we know where it is. We have signed a lease contract. And then we have a lead time of 9 to 12 months with Versatel and this -- then this site is connected to the fiber optic network. And in the meantime, we can also build the antenna carrier, no problem. If we -- as we speak of co-locations, then our partners have the right to move the sites. So even though we determine them, the co-location partner, if they notice that things aren't working out for some reason because the lessor doesn't agree, the radiation measurements aren't all right. The statics are more difficult than anticipated, the original network operator wants to modify the site, then this can lead to time delays. It can also mean that we are allocated replacement sites. And if we build the fiber optic cable at the same time, it actually goes to nothing because we go to the original site but not to the replacement, and that makes it difficult for us to have fiber optic cable in place in tandem with the site. That will change. Once we progress a few months, this will be more well aligned, but now we have this gap. In time, this will be better harmonized. Going forward, we will have, at any given time, 1,000 sites that are ready for operation -- ready for installation but not operational yet, but that will continue to be the same number going forward. But now we have a larger number that is ready for installation, not ready for operation. But I think you can see that we're solving the problem now because we -- once we have the sites ourselves and we can install the antenna and put the fiber optic in place. But the problem is to have the locations and will take a time before we can actually get this aligned, get this up and running. Let me answer your other question where you dealt with the question of our partner canceling out of certain sites. We actually do have an agreement or a clause for compensation payments. But every agreement will also have a cap on compensations that are always proportionate to what we would have normally had to pay. We, however, believe that our partner will not be able to rely or to refer back to this cap because they cannot exclude deliberate action and therefore, we will try to recover the full amount of damage caused. However, I can promise that this will be successful. We have clear contractual compensations. We will be able to get those. But whether we come beyond that, that is something that we'll have to see. And I would like to negotiate this out of court rather than going through long legal processes because we would like to continue over the years with this partner. We can also see and acknowledge that the partner is striving to achieve their goals. They've changed their -- some of their shareholders, and we'd like to acknowledge and honor that rather than destroy our business relationship now. But we'll do our best to impose this.

There is a further question by Ulrich Rathe, Societe Generale.

Just a swap of topics. You said that in the second half of the year, you expect a 4% service rate growth. Why are you optimistic that it goes from 0% to 4% in the second half of the year? And then I wanted to ask a second question, a follow-up question to Steve's question. The way you described it, it means that in principle, as soon as you have this national roaming available from Vodafone, you can speedily have the entire basis on the new deal. The way it sounded it's like a flick of a switch, but it's really a question of months or years, correct? Then a final question or clarification. You make a strong distinction between the 4G and 5G agreements. Can you give us a ballpark figure of what the share of these two different technologies are in your customer base?

Let me start at the end again. I think the distribution between 4G and 5G is about 50-50. Can we say that? Roughly ballpark figure? Yes, around about 50-50. And as you said correctly, once we're in national roaming, the migration is really the switch of a button. Whether we do that actually is a different question, whether it might be better to do it step by step. That's a different question. But it's only a data line. Our core network is connected to the Telefonica core and will then also be connected to the Vodafone core and the antennas or the SIM cards are inserted to go to 1&1 network. That's what the profile says. What the 1&1 network means at any given time, and if the 1&1 network isn't available -- we'll go to the 1&1 network. And if it's not available, ask for Telefonica network or ask for a Vodafone network. Well, this instruction can be changed overnight. So our connection to the two cores won't be -- won't change, like the one will empty out, the other one will fill up, that's it.

Another question, a follow-up question. you said that it's very difficult to move from national roaming with Telefonica to national roaming to Vodafone. But the way the technical process looks like for an amateur, it would be easy to switch from MVNO to national roaming with Vodafone because the profile is in place, flick a switch and that's it. Or are we talking of a 3-year migration or whatever?

To migrate the profile is something we can do at any given time. What is limiting factor is the capacity of caller ID porting. This is handled by an operating company, and we agreed with them that we can migrate 50,000 caller IDs per day. So as soon as we switch our profile to the new network, we also switch our phone number to the new network, and we have to do that in order to move from a Telefonica profile to a 1&1 profile. But once we've done that, we don't need to port the phone number anymore. So at the first -- in the first step, we're limited to 50,000 per day. Maybe it could be 20,000 to 30,000 per day, I don't know. Nobody ever tried to transfer that many phone numbers at a time. But once they're switched over, we can handle larger volumes depending on the technical capabilities, but we no longer rely on the supporters -- third parties for that step. Concerning the first question, what makes us so sure that we will achieve the 4% increase in service revenue in the second half of the year to arrive at 2% for the year as a whole? First of all, we focus very much on economic viability in the first part, half of the year. So we have less aggressive offers in the market in the first half of the year, just like the competitors, and that has an impact because the high-margin contracts will only become operational in the course of the year now. And also with technology changes, we implemented price increases as well, where we usually changed our offering with identical prices, and that has also had an impact in the first half of the year in realizing the fee structure change. But this will only have an impact in the second half of the year because this is when these contracts will have an impact. And also for the second half of the year, we expect a larger growth of participants due to the season. In the first 1.5 years, we concluded with an additional 180,000 contracts. And in the second half, we expect 300,000 to 320,000 new contracts. And also for landline connections, we expect a slower decrease in the first half of the year, which will be very important because there is a huge impact on service revenues. So as Mr. Dommermuth said, we expect stabilization here and even a slightly positive effect in Q4.

A couple of questions. First, we haven't really talked about the sort of market as a whole. Can you -- we've been through a lot in the last 2 or 3 years with COVID, cost of living and inflation. Can you just sort of give us your sense of German consumer behavior or any major changes you've seen in the last 6, 9 months? I mean, you just discussed the fact that you may be a little bit more careful in the first half on pricing. And maybe just talk about your confidence in the second half that German consumers or maybe where prices are a bit better in the second half than the first half, that would be great. And then just on the sort of migration side, I mean, have you had a chance to kind of compare the Vodafone and O2D networks in terms of coverage and capacity. And do you have any concerns that there will be areas where the Vodafone network is maybe not quite as good as the O2D network or vice versa where you might run into problems where customers are migrated and suddenly their service deteriorates. Is that something that keeps you awake at night? Or are you confident that for the vast majority of customers, there won't be an issue?

Well, concerning the network, of course the Vodafone network is a bit better than the Telefonica network. It has caught up a lot over the last few years. But when I see in the MVNO agreements that we have, customers can move from one network to the other. And when that happens, it usually is a migration from the -- from the O2 to the Vodafone network. If you have a lack of coverage, then you can more easily heal this by transferring somebody from an O2 network to the Vodafone network. So I think we will have a better situation than -- rather than a poor one. Now concerning consumer behavior in the face of inflation, it's very difficult to speak for all consumers. I can only say that we increased our prices and that we've also seen that with other market participants and other customers have accepted it. But we're talking about small figures here. If you pay off a mobile phone agreement plus a mobile phone, whether you pay EUR 42 instead of EUR 39, if the others follow suit, then there's no reason to make do without the mobile phone. You can dispense with other products, particularly if prices go up. But we don't see customers -- in the case of mobile phone connections or broadband connections that they dispense with this product. As long as we are well positioned compared to our competitors, we don't see any problems.

Martin Hammerschmidt. Question on the customer development. We have seen in Telefonica that the net adds over the past quarters have grown considerably. And you see a drop. Where and how do you explain that gap from Telefonica on your network quite strong and on your Telefonica network not as strong in operation to Telefonica?

First, I don't know about that. I don't know the Telefonica figures, sources and the mix of contracts. I don't know if they have one in the low budget contracts or whether they have less churn or the partners did more. Telefonica gives us overall figures only and no details. So we can't say what was better or worse there? What we have to see, that in the Telefonica growth includes our growth and in addition, our allocation from all Vodafone contracts to Telefonica contracts. So for many, many years, we have been moving contracts from Vodafone to Telefonica, that doesn't change anything in our statistics. It's always a mobile contract, but in Telefonica contracts it is visible because they get more customers. For example, if we have 100,000 new customers, and take 100,000 out of Vodafone to Telefonica, our statistics like -- looks like 100,000 new customers, in Telefonica statistics, it's 200,000 new customers. So this is something that you need to factor in, but I can't give you the detailed figures on that.

Ben Rickett from New Street in English. I had a question on your sort of longer-term subscriber development. So you're going to add about 600 mobile subscribers this year. Do you expect to do a similar rate of additions in future years? Or does that accelerate as the network builds out. And I'd also be interested in -- I don't know the extent to which those subscribers -- your current subscribers are rural and urban. But do you look to grow the base more in urban areas where you will have network coverage? And then I had a second question. I think last month, BNetzA announced they were assessing competition in the mobile network ahead of the spectrum allocation. What impact do you think that -- what are they trying to achieve with that review? And how do you think that will impact the spectrum award? And do you think they will find that there is sufficient competition in the market?

I don't know what that study will be -- have as an outcome from the agency, but the network operators shut off against service providers as of today already. You see that, for example, that city networks or regional networks do -- cannot offer mobile services. That's not because they don't want to, but they don't get any contracts offered that would allow them marketable prices. So this is why I think it's important to be available in this area. If you want regional competitors besides the fiber optic network that they build, they can offer convergence products that offer a mobile component. I think that's going to be important for the future competition. But that's my personal opinion. I don't know what the outcome of that study is going to be, but my personal opinion is that, that would describe the German market situation and that the network operators are quite restrictive there. The growth that we see this year, we see 600,000 mobile phone customers. That's quite right. In our business plan for the next years, we rather expect a drop in growth. We didn't build the business plan to say, okay, EUR 600 million, EUR 700 million, EUR 800 million, and so on. No. We rather expect that number to drop. If it doesn't happen, even better, but we didn't design it that we base on an increase of new customers that we need in order to fulfill our business plan. And the question was also whether, with our own network, we focus on urban areas. And yes, of course, I would do that. If there is an opportunity to do that, I would focus sales activities in the city areas where we have a good coverage because that is where the net utilization is of outstanding interest for us there. I think we can have one more question and then the time that the colleagues need to prepare as well.

[Indiscernible] AP Capital. A brief question only on the fixed wireless access. You've said that you have got initial experience since the beginning of the year, if I recall it quite rightly. Could you give us an impression on what that is? And what opportunities do you see in DSL access from the contingent model and switch that to fixed wireless access?

Well, first of all, we can say it worked well. The customers can have the product. We have lots of broad band, but the network is still empty. So we have to see whether we have capacities available. In that case, it would make sense to sell wireless access in a cell. And if that cell is full, and we need to put lots of effort in increasing the capacity, splitting the cell up, it's not as full. Technically, it would be possible. And Telekom has said that -- in the second -- I've just said we have won the appeal code by building the most modern 5G network. But we lost already. Telekom said you can't say that your fixed wireless access is a replacement of the landline because it doesn't do landline phone. That's quite right because when we launched it, it wasn't possible, but now we can do traditional phone calls as well, so it's a full value replacement product for landlines. And how we do it, I can't tell you as of today, but functionally, it works very well.

Thank you very much. On behalf of the Board, I wish you a wonderful afternoon. Also in the meeting of United Internet following up. And there is time for a cup of coffee, a little chat and see you again. Thank you. [Statements in English on this transcript were spoken by an interpreter present on the live call.]