Good afternoon, everyone. My name is Juan Gaitán, Director of Investor Relations at Cellnex, and thank you for joining us today for our Q1 2025 Results Conference call. Today I'm joined by our CEO, Marco Patuano, and our CFO, Raimon Trias, who will go through the key highlights of the period, and then we will open the line for your questions. If you wish to ask a question, please press the *5 on your screen to raise your hand. Without further ado, over to you, Marco.
Thank you. Good morning and good afternoon, everyone, and thank you so much for your time. I started with the key highlights of the first quarter 2025, which have been marked by consistent execution reflecting our commitment with our objective. Let me remind that our reported numbers are impacted by a change of perimeter, and there is no contribution for Austria this year and Ireland only contributed two months in 2025. On a like-for-like basis, the quarter has been characterized by solid performance across key metrics.
We have achieved remarkable growth, with revenues reaching EUR 964 million and representing an organic growth of 6.3% and our EBITDAaL reaching EUR 566 million, implying an organic growth of 8.7%. Once again, we delivered on efficiencies, on which Raimon will give you more details. In Spain, we are implementing a voluntary redundancy plan encompassing around 200 people, which is another example of our efforts to continuously optimize our operations, focusing exclusively on our core activities and improving our performance. A notable achievement has been the successful completion of the sale of our operation in Ireland.
This strategic move allows us to further streamline our portfolio, focusing on core markets where we can drive the most of the value and generate financial flexibility. From a capital structure standpoint, we have secured a new syndicated loan for EUR 625 million for refinancing purposes at very competitive terms, even in the current environment. Another highlight comes from the execution of our equity swap and the progress on our share buyback program, which started upon the completion of the Irish deal. As of the 2nd of May, we have almost completed the program, w e have acquired shares for approximately EUR 755 million.
This transaction reinforced our commitment to returning value to our shareholders, and as we shared on previous occasions, EUR 800 million constitutes our new floor for a shareholder remuneration perspective from 2026 onwards. Additional disposal may generate additional financial flexibility. In summary, we believe our journey today has been marked by the definition of a clear roadmap, a solid commercial, operational, financial performance, and a consistent and disciplined execution. Looking ahead, we are reiterating our guidance, and we will continue showing an undeterred commitment with our promises.
Let's move to slide five. Here we just want to illustrate our share price performance compared to the recent evolution of our key financial metrics. In the 2022-2025 period, using for this purpose the market consensus, we increased our EBITDAaL by 36%, our recurrent levered free cash flow per share by 46%. We reduced our leverage by 1.3 times of the EBITDA, and we started a meaningful cash shareholder return of EUR 800 million. In the same period until today, our share price appreciated only 10%, showing a very high correlation with macro factors regardless of our solid business fundamentals.
We are a leading towerco with strong and secured growth backed by the largest backlog in the industry and with the longest contracts with the clients. Slide six. The current macroeconomic and geopolitical situation is clearly impacting the performance of the financial markets, but I believe that the European Digital Infrastructure sector has a unique opportunity to demonstrate again its resilience in complex scenarios. Compared to other industries, Digital Infra is a defensive asset with a proven ability to withstand economic challenges. We have analyzed different potential consequences that, in our view, could impact the sector in the event of a tariff increase, and we believe that we are impact-free.
Let's move to customer behavior. Cellnex's strength sits on its long-term B2B contracts with its clients with a complete European focus, ensuring a stable and predictable revenue stream. Inflationary concerns: Cellnex's revenue model is robust and visible, with the vast majority of our revenues adjusted to inflation or fixed escalators. Supply chain impacts: Cellnex's strategic sourcing ensures that our operations and our investment will remain unaffected. Recessionary fears: The telecom sector is seen as a defensive industry due to its reduced exposure to variable revenue streams. Cellnex, in particular, has outperformed the broader market during crisis periods, such as the COVID-19 pandemic.
Additionally, after assessing the long-term expected evolution of the effects, the interest rate dynamics, and the quality of our balance sheet, we expect no impact. Moving to slide seven, we want to provide a quick update of our share buyback program announced earlier this year and which total today is almost completed. As of the 2nd of May, we have already executed 93% of the program at an average acquisition price of around EUR 33 per share, and we are expecting to complete the program in the coming days. As I already shared, we believe our stock is undervalued, so this moved us to consider a share buyback as the most delicate way to remunerate shareholders.
Finally, I would like to reiterate our financial outlook for 2025 and 2027, with projected growth across all key metrics thanks to the attractiveness of our business model and the visibility it provides. We demonstrated our ability to adapt to changing reality. We've shown the capacity to consistently execute our strategy, and we are creating a solid track record delivering operational and financial results quarter-after-quarter. I hand over now to our CFO, Raimon, who will provide additional details on the period.
Thank you, Marco. Good afternoon, everyone. Please allow me to provide during this section a bit more granularity on our financial and operational metrics for the period. Let me start reminding you again on slide 10 that our numbers this quarter are impacted by the change of perimeter, as there is no contribution from Austria this year, and Ireland has only contributed for two months. This table gives you a cleaner comparison as we are removing this effect from our numbers. On the left, you can see our actual reported numbers. Then, if we exclude the revenues and the results from Austria and Ireland, we got the proforma results.
Finally, on the right-hand side, you can see that the organic 2025 proforma with very strong growth and revenues and at EBITDA level with more than 6% and close to 9%, respectively. Moving to our financial performance on slide 11, our revenues have increased 6.3%, and our adjusted EBITDA has grown 7.7% compared to the same period last year. Both growth rates are on an organic proforma basis. Our EBITDA has increased 8.7% on the same basis, further highlighting our operating leverage and our ability to optimize our operations, enhancing profitability.
In terms of cash flow generation, our recurrent level of free cash flow amounted to EUR 351 million, and its growth will accelerate throughout the year thanks to the normalization of cash items below EBITDA, as I will explain later in a bit more detail. Moving to our key operational metrics, we have added 1,216 new sites in the context of our build-to-suit programs. Our gross collocation reached 1,109, and in the context of efficiencies, we have undertaken 887 land site actions backed by the creation of our land code, demonstrating our commitment to optimizing our assets and driving operational efficiency. As you will see later, we could go on improving our lease cash out per tower.
If we move to slide 12, on the first quarter of 2025, there has been another quarter of consistent commercial performance with OPEX growing 4.3% compared to the same period last year. This is explained by the progress made on our build-to-suit programs, which represents 2.5% of the growth coming primarily from France and Poland. In France, as you can see this quarter, we have had an acceleration of the build-to-suit program, allowing us to accelerate growth within the year. Net collocations are contributing 1.8% in the period.
It would be 2.2% if we exclude the impact of the MasOrange impact in Spain. In the first quarter of 2025, net collocations have been lower as a result of the already announced contract with MasOrange in Spain. As a reminder, this contract gives our client network flexibility in the short term in exchange for a single longer contract until 2048 and additional services to be provided by Cellnex. As per this agreement, no impact in revenues is expected until 2026, despite the churn that we see in terms of ops during this year.
If we move to the next slide, we provide you a clearer picture of our performance in the period. We have excluded the impact from change of perimeter. We are providing here our organic revenues bridge on a proforma basis, excluding Austria and Ireland from the first quarter 2024, and from that base, building our organic revenues for the first quarter 2025, which have grown at a robust 6.3% with a well-balanced mixed growth linked to CPI, collocation, and build-to-suit.
On slide 14, we have followed the same approach of adjusting the perimeter to look at the specific performance of our different business lines. The tower segment grows organically 5.8% on a proforma basis. Fiber and connectivity grew 24%, mainly due to the project with Bouygues in France. DAS and ACTIV grew 5%, and broadcasting services grew 3%. As you can see, we keep having a strong and well-diversified growth on all our business lines.
If we continue to slide 15, as part of the strategy set out at our Capital Markets Day and our continued focus on industrial excellence, we have the objective to become operationally more efficient, rationalizing assets, optimizing our cost base, and improving the group's overall productivity. Here, we're stripping out the contribution from Ireland and Austria to provide a cleaner analysis of the performance on our different core items. On staff, it is worth highlighting our ongoing efforts to streamline operations and improve efficiency. This reduction in headcount is part of our broader strategy to optimize resource allocation, enhance productivity, and focus on core activities.
The numbers today do not include the impact of the recently announced redundancy program in Spain, which will be implemented over the 2025-2027 period. This plan will also result in discontinuing certain operation and maintenance contracts with an associated negative EBITDA contribution. As you will see later in the P&L, we have booked this quarter a provision to execute the redundancy program, although the associated cash impact will be seen during the next few years.
On repair and maintenance, following the positive performance last year, we continue to see good progress in the period, with costs flat on a per tower basis, meaning that we are saving to compensate inflation, and the evolution of our general expenses translates into a 2.3% decrease on a per tower basis. Finally, on leases, we have been able to reduce the cost 2.3% per tower. Our land acquisition plan is picking up speed, deploying capital at very attractive returns while securing strategic locations for the future. Additionally, our efforts in renegotiation and cash advances are well on track, further contributing to the optimization of our operations.
In the first quarter 2025, we have invested 13% more in land acquisition and efficiency CapEx compared to the same period last year. The results of the strong growth of our revenues and the continuous work on the cost reduction can be seen on slide 16. Our efficiency measures are helping to first absorb inflation, second, offset the incremental costs associated with growing perimeter, third, improve our margins. Our EBITDA margin is 83% versus 82% last year. Finally, generate operating leverage. A 6.3% growth in revenues translates into almost 9% growth in EBITDA, as Marco mentioned before.
If we take a look at our debt maturities on slide 17, you can see that there are limited maturities left in 2025, as we have repaid debt with the proceeds from disposals and refinanced other maturities via new bank financing, particularly a EUR 625 million term loan at very competitive terms, as you can see. We keep working on extending the duration of our debt and flattening the maturities over time to have an even more robust and well-designed capital structure. Our liquidity remains high at approximately EUR 4.7 billion, with our debt being approximately 80% fixed with short-term maturities that have been managed.
Our average cost of debt will only marginally increase in the upcoming years. Last but not least, we have been able to execute an EUR 800 million share buyback whilst maintaining our commitment with the rating agencies. Finally, on slide 18, you can see that we are reiterating again our 2025 outlook whilst providing visibility on the expected phasing of our key financial metrics during the year. If we focus on the bottom of the slide, it is important to understand the phasing and seasonality of our cash items below EBITDA.
The first half of 2025 will be more intense in terms of lease and interest payments due to the structure of our contracts and the coupon schedule, as it already happened in the first quarter 2025. Additionally, as you know, last year, we did significant work to optimize our working capital. We ended 2024 ahead of our expectation in terms of improvements, so this impacts the working capital slightly on the first quarter, but we expect this cash item to turn positive during the second half of 2025. We have already identified measures to keep on improving our working capital. Despite timing effects in the period, we are in a position to reiterate our guidance, and with this, we now remain at your disposal to answer your questions.
Thank you so much, Marco, Raimon. Now, the first question comes from Maurice Patrick from Barclays. Please go ahead.
Yeah, good afternoon, guys. Thank you for taking the question and the call today. If I could ask a question about Zegona and Vantage. Sorry to go on the old subject, but investors do ask a lot about it. On the last call, when asked about it, Marco, I believe you said it was difficult to replicate the network due to the number of sites and possible gaps in the network. Now, Zegona is saying they think it's not only possible just to move the equipment, as in they can move it, but also many of the existing tower companies are happy to facilitate it.
There's obviously lots of tower cos in Spain. Spain is a special case in point, and it is a rapidly evolving landscape. We're keen to understand your thoughts on that. And just a small clarification. In the slides, you talk about the churn. I think it was 733 from Orange, 373 from others? Can you just let us know where those 373 is from? That'd be helpful. Thank you.
Yeah, sure. On the Zegona case, honestly, I have nothing really different from the past to add. It is a bit more of the same. Zegona is trying to force the possibility to move out from the contract. They still have time. The contract, as we understand, expires in 2028, so they still have time. They are trying to find alternatives. I am still a bit skeptical that the possibility of moving entirely the antenna that are under the all-or-nothing umbrella can find a new house, but let's see. I did not honestly change my view.
On the second part of your question, approximately, it is bits and pieces here and there. It is not a single country that makes the big number. If you take the previous quarters, a number between 350 and 450 is quite usual, quite normal. I would say that this is a specific quarter. It's mostly Italy that we have seen some churn. Also bear in mind, remember, Maurice, that branch sharing is now being classified as a physical PoP.
Also in the context of the different combinations that we are seeing in Italy, we are seeing this quarter in Q1, some disconnections of these branch sharing PoPs. I would say, yeah, that the main explanation comes from Italy, but also given that we are removing branch sharing PoPs, the associated revenues are also quite modest.
Yes. If I made the same exercise as you did, I went backwards quarter-after-quarter in order to see what was a normal churn rate during the different quarters, and it stands normally between 350 and 450. It is pretty normal.
That makes sense. Okay. So it's quite normal to see churn. We just do not normally see it. It's only because it's large this quarter you're calling it out. That's helpful. Thank you.
You're welcome.
Next question comes from Andrew Lee from Goldman Sachs. Please go ahead.
Good afternoon, everyone. I had a couple of questions. The first one's just there's press speculation today suggesting that you're into the end game in your Swiss asset sale process. I don't expect you to be able to speak too specifically about that, but just wondered what that process has informed you of on the attractiveness of selling infra assets at the moment to private investors.
I haven't heard mention of your French data centers or the Dutch business recently, but kind of any kind of broader thoughts you could give us on the merits of selling assets at the moment, be it for valuation premium or simplifying your role. The second question, just irrespective of the Swiss sale, should we expect that the next shareholder returns update you give us will be at the end of the year in terms of buybacks and dividends and not before that? Just thinking about kind of how we should think about your shareholder returns profile? Thank you.
Yeah, sure. Let's take the disposal or the asset rotation more broadly. In our case, we have the French data center in which we entered into a bilateral conversation, which normally tends to be fairly straightforward. On the French data center, I would be surprised if it derails. You never know because ultimately what we made clear or very clear is that we are not forced sellers. Not being forced sellers puts us in a better position than before.
I do not want to say that before we were forced sellers, but the target of the deleveraging and the target of accelerating the shareholder return was quite stringent. Today, if we get the price we want, we do it. If we do not get the price we want, we do not do it. On the Swiss, the speculations are saying that we should receive in those days some information about what is our market testing, which is correct. Between now and the next week, we should receive what are the non-binding offers for an asset like Switzerland, and more or less the same as I said before.
If the price is interesting, we will go to a phase two. Otherwise, we will drop it. I can't tell you at the moment, give you more color on this, but for the simple reason that I don't have more color on this. As soon as I have, we can comment on this.
I think on your second question regarding the TSR, as you know, we committed to EUR 800 million this year that we have already done through share buyback. We said that next year is going to be also EUR 800 million. That's going to be the floor. If there is any other asset rotation this year, we might consider extending the shareholder remuneration. It is difficult to give you any kind of timeline on any change on that from what we have already committed because first, we need to see if there is any further asset rotation.
Thank you. That's really helpful. Can I just have one quick follow-up just on your answer to Maurice's question? Are you able to give us the exact amount of MasOrange disconnections that you had in the first quarter and what you'd expect in the second quarter? I'm not sure we're able to calculate that from.
I think on the slide 12, you have the churn of MasOrange in the first quarter has been 733.
That's all pure disconnections from that contract, as a result of the contract.
Correct.
That is correct. I think those are disconnections expected, but also with no associated loss in revenues in 2025.
Yeah. To make clear, the way we discussed with MasOrange is to decouple the industrial or, if you want, technical flexibility we're giving to them to the financial impact of this technical flexibility. So if they want to start tomorrow morning to redesign the network, to move assets, to move antenna, please do it. We designed a cash profile of their commitments that gives some flexibility this year and the next year, but it's not a sort of a pay-per-coup. It's not a one-to-one. You move an antenna, you get a discount for the antenna you move.
Thank you very much.
This was very much appreciated by the technical team of MasOrange because they can start immediately doing things.
That's great. Thank you.
Welcome.
Thank you, Andrew.
Next question comes from Roshan Ranjit from Deutsche Bank. Please go ahead.
Great. Afternoon, everyone. Thanks for the questions. I've got two, please. Thank you for providing the slide 18 on the phasing. Is it possible to get a sense of the absolute kind of working cap number for the full year? I think previously you said small, positive, neutral, and also your thinking around the interest component, bearing in mind the buyback, the EUR 800 million buyback will be completed.
Secondly, maybe just going back to some of the previous questions, 93% done of the buyback should be completed by the end of this month, as you said. What was rationale for executing it so kind of quickly, obviously maximizing the lower share price? Is it with a view to have that flexibility towards the end of the year if you wanted to do something different or something extra? Sorry. Thank you.
Okay. Raimon will take the first. I'll take the second. Please, Raimon.
On the working capital, as we explained, last year, we put a lot of focus on the working capital, at the receivables, at the suppliers as well, trying to first reduce overdues, try to improve payment terms from customers, but at the same time, try to improve payment terms with suppliers. We did a great exercise, I would say, from June till December, and specifically in the last quarter when we improved a lot. At the end, that even had an impact compared to this first quarter. Why?
Because we even managed to achieve some connections in the last quarter that we were expecting probably to have in the first quarter. This quarter, we have had this negative EUR 20 million, but we have identified already a lot of measures to be implemented still within receivables and within the supplier side. We still have overdues to improve, and we are still putting a lot of focus there to keep on reducing. What we are planning is that we are going to finish in line with the guidance that is positive in terms of working capital for this year.
Good. Your question about the accelerated path of the buyback, the share price, the sense of the chart I made showing our medium-term improvements and the dynamic of our share price give you the sense why I consider that it was a good idea to try to accelerate as much as we could our share buyback in the first part of the year. It's still quite unpredictable what is going to happen with the U.S. rate. Possible a decrease in the rate, more difficult to predict the calendarization of this reduction.
Honestly, the only explanation I can give for the delta between our improvement in the industrial metrics and the improvement in the share price is because of the macro factors that are representing a headwind on our share price. This is why we decided to accelerate and to try to make it as soon as we can.
By the way, this is the reason why we also performed an equity swap that gave us another EUR 550 million of shares that we can buy at the price that we already fixed and predetermined in Q1 this year, between Q1 and, no, it was Q1, in Q1 this year, which would be, again, very critical. Now, going forward, if in case we have further flexibility for making something more, we will immediately look to the share price. If the share price is still convenient, we will go for that.
There was a pending question, Roshan, I think, on interest for 2025. The figure that we are contemplating is around EUR 400 million. That includes the incremental impact from the incremental debt associated with the share buyback program. Despite the, I would say, the heavy concentration of interest payment in Q1, we are not changing our view for 2025.
That's great. Thank you very much, guys.
Thank you.
Thank you so much.
Next question comes from Emmet Kelly from Morgan Stanley. Please go ahead.
Yes. Good afternoon, Marco, Raimon, and Juan. Thanks very much for taking my question. My question comes back again to slide 18. That is a super duper helpful slide for looking at the phasing of leases, tax. My question is actually really there on the phasing of EBITDA growth, please. I think I probably need a little bit better eyesight to figure out what is going on EBITDA Q1 to Q4 in your slide. If I just look at the numbers, EBITDA fell about EUR 65 million quarter-on-quarter between Q4 and Q1. Just looking like the last couple of years, EBITDA has increased sequentially between Q4 and Q1.
I know there have been a few kind of exceptional items in there for Q4. I know there was the loss of the Austrian and Irish EBITDA, which you mentioned, Marco. Can you just say a little bit about what's going on underneath the hood? I think there's something going on in terms of the engineering works that you guys are providing for telcos and maybe some other elements as well. How should we look at that EBITDA number evolving, let's say, Q2 to Q4 this year? Thank you.
Yes. I start giving you the first part, and then Raimon will complement with further details. Yeah. You touched two big offenses. One, of course, is the discontinuation of the operations in Ireland and Austria, which is fairly material. The majority of the number comes from there. We have three months not having Austria and one month of Ireland that we do not have. That is part of the difference.
Yeah. It is quite material. This is why Raimon made also the apple-on-apple. I do not remember exactly the number of the chart in which you restated all the numbers. We chose chart number 10, which I think is fairly helpful because at least it gives you the sense of what is this component. You were mentioning the engineering services, which is another part of this activity. Engineering services move together with, how can I tell you, the activism of our clients, which tended to have seasonalities.
It is 30 years that I am in this industry, and it is 30 years that I am trying to understand why engineers tend to move their feet more in Q4 than in Q1. It is exactly what happens every year that golf gives us. The point is that in Q4 2024, our client had a sort of hyperactivism in this. In Q1, possibly they took a nap. There is a bit of difference coming from there. Nothing that puts at risk the total number for the year. The calendarization, yes, it is true that the calendarization this year, or if you want the seasonality this year, has been a little bit higher than what has been the previous year. Raimon, if you want to complement?
Just to complement on the engineering services, you will see that the main difference comes from France. That is the place where we had most growth in the last quarter and then less growth this quarter. You were mentioning also, Emmet, the one-off. It is like that also in the fourth quarter, if you remember, we had this fine in Spain that we released a provision for EUR 18 million because we won the case. That impacted one of the lines of cost as well. That was explained during the end of the year.
Super. That's very helpful. Thank you both. Thank you.
Thank you.
Thank you. Next question comes from Rohit Modi from Citi. Please go ahead.
Hi. Hope I'm audible . Thanks for the opportunity. Just a couple from my side. Firstly, I'm not sure if I'm reading too much into it, but Marco, your recent comment around growth in the U.K., Netherlands, Spain, once this overall completes, and also a statement from the chairman around you're looking at EUR 5 billion CapEx investment opportunity over the next three years. If I look at your CapEx guidance in terms of implied CapEx guidance, that's around EUR 4 billion.
Just trying to understand if there's more investment that you're looking into any of the markets in the next few years, any color on that? Second, on the land aggregator regulation that you have been mentioning a couple of times in the past, how do you see what kind of risk do you see around the land aggregator issue that you're facing? Lastly, equity swaps. Just wanted to understand if you want to continue the equity swap, post this buyback, or it will probably terminate. Any view around that? Thank you.
Okay. Good. On the future growth CapEx, I think that the vast majority of the growth CapEx will stay linked to the towers. We have to end the fiber project. The fiber project in France is, I try to remember by heart, should be a sort of the remaining CapEx. I am asking my colleagues, a sort of EUR 200 million I have in mind by heart. If I am wrong, we will pass you the number, but I do not think I am materially wrong. Raimon is telling me that I am right, so I am good as a former CFO.
When I look at markets where I see possibility for having network creation, I continue to think about France. We continue to have network creation. U.K. would continue to have network creation. Poland is going to be for us an important market for making both a network creation and a network rationalization program. This is going to be quite important. The land, you have to imagine something between EUR 250 million and EUR 300 million a year. This is not a number that's easy to come to my mind. Land brings me to answer to your question on the land aggregator.
Thank you for making the question because I want to make it very clear. Point number one, I have nothing against land aggregator if they do the land aggregator. The land aggregator who have cheap funding go and buy land and then turns to us and they become the new landlord instead of being the farmer is a land aggregator. I have nothing against. I tell you that eventually it's a more professional counterparty, i t's someone that you optimize the administrative component.
So many things. Sorry, there is some noise around us. I hope it does not disturb the communication. What I'm strongly against is the land aggregator who try to have unfair return on their investment, which means they get the right on the land, then they turn to us and they try to have an undecent profit with us. This is something that I consider simply not in the interest of my clients, not in the interest of the community because we end up reducing the coverage and it's going to be something bad for the entire system.
We're going to fight this attitude as much as we can, and we will fight it hard. We're not going to let them put us in a corner. We will fight, and we will fight hard. I hope that I've been clear and polite. Equity swap. As of today, we do not see the possibility of having more equity swap. Today, we have EUR 550 million that we bought in two different tranches.
The price we made for the EUR 550 million is around EUR 32, which is very nice. The financial component of the equity swap is very well priced. It remains a very attractive price even if we wait a little bit to exercise it. We do not expect in the near future to have more equity swap.
The maturity of the existing equity swap is June 26. We could accelerate if we wanted, but if not, we have until June 26.
Thank you.
Thank you, Rohit.
Next question comes from David White from Bank of America. Please go ahead.
Yeah. Thank you, guys. Just a slightly different question. Some of the towerco's talking more about owning the RAN as part of the tower contract and providing synergies upstream to the operators, potentially sharing the RAN a lot more. I know there is already some RAN sharing in place, but I just wondered how you thought about extending the sort of traditional tower model upstream a little, let's say, into the more active equipment.
I think we're all broadly agreed that sharing telco networks is only ever a good thing, really, and provides synergies to potentially everyone. I just wondered how you thought about that evolution of the business model within your business plan. Appreciate it's a little more hypothetical right now. Thank you.
Okay, David. As you know, we already have this model in place in Poland. Possibly I'm the only one in Europe who can talk about the topic having a direct knowledge of this stuff. You get synergies if you have a network that serves more than one MNO. If you have one network which is serving one MNO, you can make it more efficient. You can rationalize. We are pretty good. By the way, we have less constraint in being more efficient than some of our clients. The real efficiency, you get the real efficiency if you have two networks, you combine the two networks, and you make a single network.
This is the way you make real efficiency because you make efficiencies every time there is a change of generation because instead of making two carpets, you make one carpet, and you have synergies on OPEX, maintenance, transmissions, everything, energy. You make synergies. I personally see more headwind than tailwind going forward in having these RAN codes for the mobile. For the mobile, it's a little bit more complicated than for the fix because you have to involve the use of the spectrum, the allocation of the spectrum, the ownership of the spectrum.
It's a bit more complicated than in other cases. To say the truth, we are working. We are learning in Poland. Hard for me to believe that in the short term, we can replicate these in other countries. The investment profile is honestly different from the investment profile of the tower code. The investment profile of the tower code is something that you put the money upfront, and then you marginally upgrade or you marginally change it during the life of the investment. The big investment is day one, and then you go for 20 years.
When you are in a RAN code, you have an investment cycle every five, six, seven years because of new generation. Now we have the 5G, and 6G is already beyond the corner. We can argue if 6G would arrive soon or late, but in the Far East, they are really already starting really talking about 6G. In our case, it's something that it's not easy to imagine Cellnex heavily invested in RAN or, as we say, RAN as a service. Having said that, where we are, we are not unhappy, and we are learning a lot. It is a good business for us. Possibly, I see RAN codes being created and being not necessarily integrated with tower codes.
Okay. Thanks for your answer, Marco. Nice to speak to you.
Thank you. Thank you, David.
Next question comes from Ondrej Cabejsek from UBS. Please go ahead.
Everyone, thank you for the presentation and for allowing me to ask questions. I've got two questions. One is a bit higher level, just on your build-to-suit contract and how well protected they are legally in terms of future growth. Because obviously, when you have a potential merger in one of your markets, if there's a party merging that's got an outstanding build-to-suit contract with you, then the future growth from the sites that you have not yet built, how well is that legally protected in the sense that this is or was reflected in the multiples that underwrote the deal in the first place?
We always talk about the current perimeter of revenues being very well legally protected, but how is this the case for revenues that are not yet kind of in the ground in terms of towers not yet built? If you could elaborate on that piece. The second question is more of a follow-up just on the MasOrange situation in Spain, where obviously you've seen slight decommission as you report.
You say no impact thus far on revenues, but I guess we're correct in assuming that over the next five years, there will be a dip somehow before you recoup revenues for the new services that you highlighted in the first release maybe three, four, five years down the road. Is that still kind of the expectation? Thank you very much.
Okay. Sorry, I hope I understood perfectly correctly because it was a little bit disturbed, the line. The BTS, in case the BTS agreement stays in place in case of a merger of two clients, it is not necessarily true, but this is an easy discussion to have with the merging entity because if they really continue to need it, we are happy to maintain our CapEx allocation.
If they do not want it, they just have to tell us, and we do not put the CapEx and possibly we renounce to a little bit of CapEx with this client, but we can reallocate the CapEx on growth with other clients. Today, the problem is not that clients are not asking us to make more towers. The problem is that we are disciplined. We have a maximum framework of CapEx that we want to put on the field every year. Sometimes we say no.
If I cancel someone on the right hand, possibly I can put the money on the left hand, but believe me, it would be an easy conversation with a merging entity because it is useless to force someone to have something that they are not interested in. It is not okay. The MasOrange case, yes, you are right. It goes down, but it is mid-single digit in absolute terms. Let's say EUR 4 million-EUR 5 million.
When we recover, we recover a little bit more because we tended to recover including the financial effect of the displacement, of the time displacement of the two events. You reduce first and you recover later. You cannot recover exactly the same amount. You have to recover a little bit more. It is something that has been designed together with the client. The client is perfectly aware, and it's a hard commitment. This is important. It's a take-or-pay commitment. Of course, nobody likes to pay for nothing. I assume that they will use the commitment for which they're going to pay.
Thank you, Marco. If I can follow up on the first point around build-to-suit, please, because I guess my point was, yes, you can allocate capital elsewhere, but if you do exit or if your client exits from a build-to-suit program, you are losing future growth that is currently expected, and that expectation was built into the transaction multiple if I'm thinking about this correctly. Is that the correct kind of way to think about it, or?
Yeah. Let me try only. Bear in mind that our build-to-suit, the economics of our build-to-suits are more similar to the original M&A transaction rather than organic, if you like. Okay? If you compare the average cost per tower, it is very similar to the initial M&A transaction, and also the associated EBITDA per tower is also very close. Again, the economics of the build-to-suit are very similar to the M&A. Yes, clearly, if we do not continue with some of the ongoing build-to-suit programs, we do not see the associated EBITDA, but at the same time, we save on the CapEx. I do not think that any discontinuation would change the original return of the project.
Yes, I agree. You have to imagine some of those contracts which were including build-to-suit as a deal with a component which has a forward execution. That is it. That's the way you have to read it.
Okay. The equivalent terms, and it's I guess the thing that I was missing. Thank you very much.
You're very welcome.
Next question comes from Ottavio Adorisio from Bernstein. Please go ahead.
Good afternoon, gentlemen. I have a couple of questions on my side. The first is a follow-up from what Marco said about the land aggregators. It's clear you like them if they're fair. You don't like them if they are unfair. I just want to see if they are unfair, which are the tools for you to avoid a contract being terminated? Because from memory, your average leasing contract on the land is significantly shorter than the average contract on the towers you have with the clients.
If basically that contract reaches the end, they expire, and it's not been renewed, how are you going to cope with this, especially if the land aggregator will have a significant amount of land where you can't actually build anything around so he can provide a service? The second one is a follow-up on the Orange, but more on the U.K. In the U.K., you also have a similar situation whereby you're given some flexibility to the client's Vodafone. It's a different sort of agreement, but you never really disclose how the agreement works.
Should we expect what's in the Vodafone, some dip on the revenues in the short term with a compensation medium-long term, or the revenue cycle is slightly different? The third one, it's just a follow-up. When this tool we won, when you said basically on the M&A that the BTS is a similar business model, from memory, you don't build any towers. The towers are built by the operator, so you just buy the towers. That's the reason why either you do on BTS or you do on M&A, it's exactly the same transaction. Is that correct?
Okay. Yes, just to finalize. That is correct.
It's partially correct.
Partially correct, but in terms of the economics, yeah.
Yeah. It's partially correct because we build approximately one-third of the towers that are in the CapEx programs. Two-thirds are built by the operator who sells us with a similar model. One-third we build, and we have the full responsibility, but the fact that you build or someone else builds doesn't change very much the economics of the program. What changes is, let me say, the activity that our engineering department has to do. On land aggregators, land aggregators, point number one is our contracts are the duration of our contracts is longer.
The average duration is longer than what you could expect. The average duration of our contracts is more than 20 years. Average. Like every average, it means that there is something that is expiring tomorrow morning, and it's something that is expiring in 50 years. The big effort we are making is it's a tremendous exercise we did in the last 12 months is putting all the data of all the contracts into a single big data lake and adding some artificial intelligence on it and starting making analysis on what is more at risk and what is less at risk and which kind of fatality has the risk.
If I lose a rooftop in the heart of Paris, it's not the same that if I lose a land in Provence. I can replicate in one case and hard to replicate in the other case. What we do is the following. Where we have site at risk, we call it SAR, site at risk. When we have a site at risk that our AI tells us that replicating it would be a nightmare, we run for buying the land immediately or extend the contract as much as we can. If a land is in an area where we can have an easy alternative solution, I cannot tell you that I am happy, but it is less urgent.
If someone comes there and wants to have a nice fight, okay, we say several bad words, and then we start building a tower in the field next to the first one, which I tend to dislike because I spend money for dismantling a tower. I spend money for rebuilding a tower, and I have to ask pardons to my client because I will give him a bad service for at least two or three months. We will never say yes to someone who wants to blackmail us.
I hope I was more clear. The orange, sorry, not the orange, the U.K. merge case, yes, it's slightly different, but it's not at the end so, so different. It's slightly different because, first of all, the timing of their network redesign has not been as quick as MasOrange. MasOrange, the day after they got the authorization, they started building, dismantling, moving.
So they've been hyperactive, and I like active people. In the U.K., they are taking a little bit more time, but the concept is not very different. There is one thing that is the capacity to move or the permit that MNO has for moving assets, and then there is the financial profile, which is more linear than what is the movement of the towers. Yes, it's a bit different. The underlying concept is not very different, Ottavio. I hope I answered your questions.
Perfect. Thank you very much.
Thank you. Thank you, Ottavio.
Next question comes from Luigi Minerva from HSBC. Please go ahead.
Yes. Good afternoon, everybody, and thanks for taking my question. It's a topic we often discuss with investors, and it's about the influence on the equity story from macro and rates. I guess the question is whether there is something you can do as a company to decouple your equity story from the rate cycle, and this may be operational actions, maybe balance sheet actions, or should we just accept that this is a long-duration asset and nothing to do? You are exposed to the rate expectations volatility. Thank you.
If someone has a magic answer, please write me an email. Do not wait until tomorrow because I would love it. I take only the part on the balance sheet. Possibly you mean what if we deliver more? Of course, we are thinking what if we deliver more. I am conceptually in a dilemma because if possibly it is true that in a higher-for-longer scenario, you should deliver more, which is unproven how long is this higher-for-longer. On the other hand, I have to consider the profile of my contracts.
I have a 20, 30-year contract, highly predictable, covered against the inflation. Every time I look at the intrinsic evaluation of our business, ultimately, it is a DCF. Now, if I deliver too much, my work goes up, and if my work goes up, my DCF goes down. It is a bit of a dilemma of the prison here in which, on the one hand, a higher-for-longer scenario suggests a lower, or if you want a more aggressive leveraging. If you look at your cash flow profiles, what suggests to you is not to go too low.
Let me put it this way, Luigi. If we will stay higher-for-longer, the leveraging possibly could take more momentum. Today, we are not really convinced that rates can eventually even go up. We are still in a macro in which every time we talk with macroeconomists, they say sooner or later this is going to go down more than up. Our idea of moving in the direction of 5.5 stays there, and then we will see.
Thank you, Marco. Appreciate your thoughts.
Happy to have separate conversations with each and every of you if you have a different view. No, I'm serious. If you have different views, there is another thing that drives me crazy, Luigi. I have not even a penny in US dollar, and I'm linked to the US dollar rate.
We have a correlation to the U.S. rate that is higher than the correlation of the American peers, which makes no sense, no?
Yes.
Probably it is true that they have a lower leverage, s o we have committed.
Not all of them.
Not all of them. Not all of them. We have committed to reduce leverage, and we're still on the way. Probably what we have seen is that our correlation that was very high has slightly reduced a little bit, but we are not yet as the others. Hopefully, as we keep on deleveraging and reaching the five to six times, we will see an improvement, but it's not going to be massive, and there will be a correlation.
Yes. Nevertheless, we are correlated to US dollar 10-year and not to the euro 10-year. I understand that our investors, most of them, invest from portfolios that are US dollar denominated. This is the obvious explanation. If you look to the intrinsic value of the company, it's like saying that I have the revenues that are linked to inflation, European inflation, and so European rates, but you make my discounted cash flow based on the US dollar rates, which is at least strange.
Perfect. Last question comes from Fernando Cordero from Santander. Please go ahead.
Hello. Good afternoon. Thanks for taking my two questions. The first one is related with the efficiencies delivery that you have already shown in the presentation. I would like to understand how confident are you regarding sustaining this unitary cost per tower OPEX control? In that sense, to which extent do you require additional, let's say, efficiency caps when talking on the OPEX-excluded leases to maintain this, let's say, efficiency level at a per tower basis?
The second question is, on the whole discussion about MNO consolidation in Europe and what would be the impact in CapEx, I would like to understand, given your knowledge on the ground, on the side, of course, how do you define the pending investments in 5G deployment in different markets where you are already present? Thank you.
With enthusiasm, Raimon takes the first, and I will take the second.
Thank you, Marco. Fernando, as you know, for the past 12-18 months, we have been working a lot on trying to improve the analytical information that we have in the company. We have developed tools inside the company to understand, first, what is the profitability and the cost on a business line by business line and a country by country.
We have divided also the P&L by what we call commercial expenses, what we call operational expenses, land expenses, SG&A, so that we are able to identify which costs, for example, are more fixed, more variable, which ones are linked to volumes, which ones are linked to sales rather than costs linked to towers directly. Based on all that, Simone and the team have been doing a lot of internal benchmarkings to understand who are the best in class in the different categories.
We have done exercises on trying to understand how many towers every person is managing and things like that to keep on doing savings. As you have seen, we are managing to keep the cost below inflation and even reducing the cost per tower in some cases. We believe that this is going to be kept during the year. That is what we are planning.
We have put targets to all our teams to keep on improving, and we have created this benchmarking activity that we meet every month among the different countries with the CTOs, the CFOs, etc., to make sure that we have the best in class, and everyone has to tend and improve to that level. We create initiatives, actions to make sure that we get there. From that perspective, that is working.
On top of that, the operations team has developed the ABC sites. We have now a P&L not just by country and by business line. We have it on a site basis, and we are now looking on a site basis to understand which sites are having the highest direct cost to see how we can mitigate and reduce those direct costs to keep on saving and improving the cost per tower on a monthly and annual basis.
Yes. To complement this, I think that going forward, you will see more rationalization projects. Rationalization projects are projects in which I have two towers too close to each other, which impede possibly both of them with tenancy one, which impede that tenancy ratio because they are too close. Possibly not only what we were calling colocation to suit, to avoid to build, but what if you already built? Is it possible to have a rationalization project?
With Simone, our COO, we are working. We are working on Poland. We are working on France, which are the two countries. Italy possibly could be another case. This is something that, let me say, a bit more midterm, but short-term can become very interesting and attractive for making the synergies. Now, your second question is 5G rollout by country or where is Europe?
Europe is a bit of a mixed situation where, not surprisingly, countries with three operators and with a higher RFU have a better network quality than others. If you take, for instance, Switzerland and Netherlands, they have very nice network quality, which means that the market is stable. For us, it is a good cash generation, but it is not so much growth. We are talking markets that are very solid, and they are going to work on special situations, the subway, the indoor, that, of course, can represent for us further growth opportunity, but it is what it is.
If I say today, what are the two cases that we are monitoring closer in terms of 5G needs? Poland is for sure number one, and the U.K. is number two. U.K., the CMA is pushing the merger to go to 26,000 ops, and all the other operators are between, let's say, 17 and 19. Hard for me to believe that if you have one operator with a quality coming from 26,000 installations, the others can stay at 17 or 18. There will be a catch-up. Mixed feeling for France. France has a nice 5G in urban areas, and when you get out from the urban areas, the situation is very different.
You have decent 5G, not even this. You have 5G on the traffic routes, the major traffic routes. The more you enter in the countryside, the more you start suffering. France has a lot of countryside. By the way, this is something that even the government, the French government, is considering. They had a first project, which was the Sande, you remember the name? New Deal. New Deal was the name of the project that was a rural project, and they are thinking New Deal 2. How to continue to push operators to improve the coverage on those.
Italy, Spain, to stay on big markets. The average quality is not bad, but then it depends if you are future-proof or not. I think that if the traffic continued to grow with the current path, they can be considered that the improvement can be marginal. If AI kicks in a bit more strongly on the consumer side, also those networks are not okay. Network okay today, I would tell you, Netherlands, Switzerland, Austria, and not so many more. The Nordics, Sweden and Denmark, Sweden and Denmark, they are really in the phase of thinking which kind of network rationalization they have to do.
Denmark, they have a relatively good 5G, but fairly inefficient. Sweden, it is a strange topography because you have the coastline in which you have to deliver a good service, and then the more you go north, the more you find Santa Claus, and it is difficult to give coverage to the deers. It is a bit difficult. Portugal, they have a relatively good network. All the operators have good networks. By the way, they performed, honestly, quite well during the blackout. They suffered. They had an interesting performance. It means that their networks are, to say the least, well designed. I hope I answered, Fernando.
Yes, many thanks for the capillarity on the answer, Marco and Raimon.
Perfect. We can finish finally the end of the session. Thank you so much for your time, and have a fantastic weekend. Take care.