Merlin Pptys Socimi Ord

Good afternoon, ladies and gentlemen. Welcome and thank you for joining Merlin's first half 2024 resource presentation. Our CEO is Matt Clemente and CEO Miguel Aguero will be going through the slides that you can see on the screen, which are also available in our website. We ask you please to abide by the disclaimer contained on it. After the presentation, we will open the line for Q&A. For those of you who want to raise questions, please press star followed by number five. With no further delay, I pass the floor to Ismael. Thank you.

Thank you, Ines. Good afternoon, everyone. Welcome to Merlin's first half 2024 results presentation. The company has had a very good start of the year with strong operating performance overall, translated in a very solid like-for-like rental growth, positively spread in all asset divisions, and sustained high occupancy levels. In offices, we have enjoyed positive release spreads, decent like-for-like growth, and while we indicated to the market that we expect the year to be flat, slightly negative as compared to last year, that meant basically going from 92.5 to something between 92 and 92.5, Our current stance is that we will end up the year flat but with positive tilt. So it's going to be between 92.5 and 93, probably closer to 93. We are experiencing or we continue experiencing good dynamics in logistics with a 4.1% like for like and a positive spread. The relatively low For recent standards, occupancy of 97.6 is a consequence of a cut-off date effect. A number of modules on existing parks have been left vacant during the last quarter, but have subsequently been re-let. So towards end of the year, we should go back to virtually full occupancy. Regarding shopping centers, they continue outperforming with a 6.4% release spread and a 96% occupancy. And what is better, the occupancy cost ratio stands at a very record low, 11.5%. In terms of financial performance, we have achieved $0.31 of FFO per share, flat, slightly positive versus last year, which is very important because, you know, the company is little by little normalizing after the effects of the sale of the VBA branches, and it's also little by little digesting the increased overhead created by our data center activity. In terms of valuations, we were positively surprised by significant uplift in data centers, plus 13%. that has more than compensated the value erosion, slight value erosion in the other three asset categories. The yields continue normalizing. Overall company yield now stands at 5.2% passing, and offices, which is the one that probably concentrates all the concerns in the market at present, is now at 4.8%, I guess, it will continue tilting towards 5% and something percent during the year or maybe next year. The financial position of the company remains very, very strong, 35.6% loan-to-value with fixed interest rates. We have no debt maturities. Well, we have one debt maturity in May 2025, which is already covered. So we have no uncovered debt maturities till November 2026. and are at present holding on 1.6 billion of liquidity of which 725 million is cash in advance of the May 25 bond maturity and the rest is ungrown trade lines. As you all know, S&P upgraded the company to BBB Plus with a stable perspective thanks to our lower leverage and the perspective of improved cash flow generation through the data center activity over the coming years. Regarding value creation, we have slightly increased the amount of non-core disposals that are either signed or in advanced negotiations. We are indicating high single-digit premium to GAV, and we expect to materialize those towards year-end as we have tried to protect company cash flow and we prefer to execute the purchase and sales towards the year end rather than in the middle of the year. Regarding best two and three, both plans have now been merged, so we will report to you on the basis of what is short-term, medium-term, and long-term so that you can expect what is coming from the logistics WIP. We have in excess of 200,000 square meters of land bank with pre-let or agreed head of terms that will materialize over the coming months. So the WIP will continue adding to the logistic revenues until 2027. I mean, just by heart, I mean, just those 212,000 square meters mean $10 million, $15 million of extra gross rental income to our logistics division, which of course will be very much welcome. Regarding the mega plan, at present we are, you know, helmets on, and what we are doing is focusing on the equipment of the operating data centers pertaining to phase one And we are also doing the preparatory works for the construction of phase two. We are, of course, working on the pre-bookings for phase two, but as you all know, and I'm sure there will be plenty of questions during the Q&A, we first need to tackle the sources and uses so that the funding is assured and we enjoy enough cash at banks to make sure we build the two data center campuses in Lisbon and the Basque Country. The licensing of those new projects is on track. In Bilbao Arazur for building number two, construction permit is expected to be received in 4Q2024. maybe I alluded to that earlier. In Lisbon, Villafranca de Xira, we have already received the construction permit as is public. We are finishing the organization works by the end of August and we will start immediately compaction and piloting of the land for basements in September. As you know, this is a very special type of soil and we need to do a lot of preparatory work. Regarding value creation of data centers standalone, we are in early stages of the appraisal uplift that this will bring, talking just about phase one. We have registered in the books around 120 million of value uplift, which is between one-fourth and one-fifth of the total that we expect just for this phase one, which should be in the range of between 500 and 600 million. So very, very interesting. As commented many, many times with many of you, very interesting secret weapon that we keep in our balance sheet in order particularly to compensate any potential value erosion that might still affect the traditional asset classes. Regarding phase two, I know that some of you do not necessarily agree with the way we account that we are keeping it at historical land cost. We will add CAPEX while we progress in the WIP and start appraising upon inauguration. This is the way we prefer to do it in order to be prudent because otherwise that could be a very important figure and I don't think it will keep a good image of our company. I am sure there will be lots of questions rephrased in very different manners during the Q&A session about funding, but I want to tackle upfront your potential queries. Basically, at present, The BOD, as recently as Friday, approved our DC business plan. That's it. So the data center business plan, both base case, which is the one you will see in the presentation, and all alternative cases have been approved by our board of directors. And more importantly, for what it means, with the full support of our two co-shareholders, and the executive management of the company. That plan requires around 2.1 billion of funding. That is also public information. You all know that. The management has been tasked with the planning and executions of the actions required in this regard. The idea remains to fund plus or minus 50-50 equity and debt with the equity being injected first. And therefore, by year end in normal circumstances, although as you know, we tend to be prudent in our forecast of when things need to be done. Any initiative requiring equity will certainly be carried out at merit level why because we wanted to preserve as much upside as possible for the benefit of all existing shareholders of the company we didn't want to create a leakage of upside downstairs no matter it is possibly compensated by the pricing of a specific capital increase you never know i mean You could be leaving some money on the table, so we prefer to do it at the mother company level. Likewise, and in the same vein, any initiative entailing a new shareholder cannot result in granting, however big, any special rights. whether of first look or co-investment or similar. So, you know, we want to be fair other than those simply attributed by law. If and when needed, the BOD will meet and decide specifically on funding. And until such moment, it is of course our duty to not speculate about any potential transactions. very different thing is to talk theoretically about funding needs in February, how can you fund this project, et cetera, than talking now because everybody knows that whatever needs to happen will happen more immediately than in February when we talked about this transaction, which, by the way, the market now knows perfectly well and is perfectly anticipated, or the market, I guess, is expecting it. So I leave you with Miguel Ollero, who will talk about the financial results of the semester. Good afternoon, everybody. We are moving now into page six of the presentation. We are going to digest a little bit, which is the financial performance of the company during the first half of the year. As you can see in terms of rents, we were reaching the 248.2 million euros, which means plus 4.4% with regards to the same period in 2023. In the case of office and logistics, they went up by 5.9% each of them, reaching 134.6 million euros of rents in the case of offices. 42 million in the case of logistics. In the case of shopping centers, there was a decrease in total amount of 1.2%. This is driven by the fact that we were disposing two assets last year, which are not generating rents in 2024. Nevertheless, as we will see later, in terms of like-for-like and release spread, it has been the greatest contributor to the company itself. So it's just a one-off thing to take care about. If we look at gross rents after incentives, I should be highlighting that incentives are lower than they used to be in the previous period. So this represents 5.5% of the gross rents, whereas last year they were 6.5%. So the company also as a consequence of reaching full occupancy, is also able to reduce the incentives within the portfolio.

In beta terms, we were reaching $188.4 million, 3.7% ahead of the first half of last year.

So in the end, all the operational performance from the financial standpoint is outstanding with regards to the previous first half. Even at FFO, despite the fact that we have been commenting that the data center division is already in a ramp-up and is already not performing from the operational standpoint at full, as it is already understood, we have been able to maintain a similar FFO within the company, which means that we have reached the 31 cents per euro per share pretty much with the first half of 2023. With regards to the EPA MTA, it is 15.11. We consider this with regards to the end of the year. We should be accounting on the 0.24 in June. The company has been already reaching TSR for the first half of the year of 1.8%. Moving to page 7, you have here devolution of gross rent on a lack-for-luck basis. It is 2.8 overall, and it should be highlighting that logistics was at 4.1% and shopping centers at 3.3%. Moving further into page 8, occupancy-wise, as Ismael was pointing out, we are in a very good position in terms of offices. The GR is evolving much better than we were expecting. That's why we are confident that by year end should be more than 90% occupancy. That should be the highest occupancy ever in the office portfolio for the company. Logistics, 97.6%, but should be as close as possible to 100% by year end, although this is something that is not common in the market to be at such high occupancy within a portfolio. And shopping centers, quite stable, 96%, in line with the So the company itself is ranked between 95% to 96% occupancy overall. Now we're moving into each asset class, and Ismael will be highlighting the main topics around the different asset classes of the company. Okay. Well, regarding offices, anecdotally, In this semester, Madrid has overtaken Barcelona in terms of occupancy, owing mainly to the relatively weak performance of Barcelona, reflected mainly in the 22-at district. The rest of the portfolio is shining, as always, but in 22-at, there are some problems of digestion of oversupply that will remain for a number of months ahead. However, the important thing is that Lisbon continues performing stellarly with 98.8% and in Madrid things have improved mainly due to the performance of the A1 corridor, which we will put the focus a little bit on it in another slide. As for leasing activity, it's been a relatively busy semester with close to 100,000 square meters transacted. Well, release spread is modest at 1.1%, but you have to take into account, as commented on many occasions, that we have been passing a lot of inflation to tenants and as such this is the nemesis of inflation that really spread. But the tone of the market and the transaction activity remains very interesting. Number of visits and leads remains strong and we are experiencing a relatively healthy market for the time being and let's see what happens in the future because you know the transformation into residential theme is so strong in Spain that let's see what happens. We like what we see in terms of stock reduction. In Loom, we have opened three new spaces with a little over 250 desks, but the lion's share of the new desks will come in the second half of the year with over 700. Little by little, Loom continues trending towards 40,000 square meters, which is the point in which we expect that beyond its contribution to the services provided to clients by the company, it will also become a net revenue contributor after specific overhead expenses for the company. What is really interesting is the ADR that we have achieved. It goes to 400 euro per desk. which translated into a per square meter basis. We have commented on many occasions that represents 1.5 times market rent, which is a very, very interesting way to commercialize space, even if it is on a relatively short-term basis. Occupancy stays flat at around 82%. You know that our policy is to introduce new offer when occupancy goes very close to 100% because, you know, we need to continue providing services to clients. Anyway, this is an anecdotal part of our portfolio. It represents around 2% of our total renting offices. So nothing really to be too much, you know, bragging about. Regarding the A1 corridor, well, you know that the company since the acquisition of Metro Afeza has been in an uphill fight to improve services vacancy or to improve occupancy on the A1 corridor. But now we are pretty happy, we are pleased to see that those measures have yielded results and occupancy in the A1 corridor has now increased by close to 80,000 square meters since 2018. That is 18 points of occupancy, full points of occupancy gain. We have taken a lot of initiatives in the area, which represents 300,000 square meters of our total stock, and including launching the Merlin Half Concept, which is, to our knowledge, is the largest business half concept in Europe, encompassing 28 buildings with more than 75 top-tier companies that are, in a way, dealt with all together by us, by Merlin, with a special app in which at the touch of a button, the users have access to things like gym, paddle, tennis, live events, food and beverage options, a shuttle service, We have even negotiated and obtained from the municipality of Madrid a dedicated bus lane that moves privately our customers from the area to a specific communications hub in the north of Madrid, which is clearly reducing significantly the use of private cars and hence the traffic jams in the area. The tenancy schedule is fantastic. We have commented on many occasions that some of our most recognizable tenants are not in CBD because, you know, an average patron in Spain, an average Spanish citizen will not recognize the logo of Allen & Overy or, you know, Goldman Sachs. Most of our best tenants are located in the out-of-town locations, including, you can see, PWC, BBA, Indra, Técnicas Reunidas, Dr. Gamble, Phillips, very, very interesting. So vacancy in the area is now slightly above 25,000 square meters, and we will continue working on it in order to make that vacancy even smaller. We are being helped undoubtedly by the proximity of Operación Chamartín. We have commented on many occasions that our bet on this A1 corridor was because at some point when the Opera Fentanyl marketing was carried out, this will be kind of first lane to the beach. So, you know, that is now starting to yield some results because the heads of space, the head of real estate of different multinationals are now, you know, little by little giving more importance to this area. In logistics, well, Very interesting performance, as always, and very interesting like-for-like growth. Occupancy in Madrid is flat. In Barcelona, it went slightly down, but it will be recovered towards the end. And in other areas, which is mainly Zaragoza and Barcelona, it has gone slightly down, but it will increase owing to a number of transactions in Seville before the end. So we expect to close the year at a very interesting occupancy level, you know, as close as possible or maybe exceeding the one we obtained last year. Leasing activity very robust with more than 60,000 square meters transacted. And South Port in Barcelona recovering in terms of occupancy now at 97.9, you know, which is pretty good, really spread negative, very mostly negative of 1.7, owing mainly to one big contract that dragged a little bit on the, it was over-rented and we had a little drag on the rent. We have recovered a little tad of stock because part of the third-party stock, the ground leases, 7,000 square meters have now moved into stock under management, so into operated stock. So very interesting also the performance of Southport in Barcelona. Shopping centers, once upon a time the Cinderella of our portfolio, now the princess. The like for like has been 3.3%. And what is more important is we continue enjoying better sales per square meter and better attendance, better footfall than in 2023. And the OCR continues going down from 11.5. You might remember that our historical averages are between 13% and 14%. But now we are at the historical minimum at 11.5%. Interesting activity in during the semester with close to 31,000 square meters. I will leave you with Miguel to discuss about valuation and debt position. OK.

Regarding valuation on page 22, valuation for this first half was pretty flat.

But if we're going into the different asset classes, it was slightly negative in the case of office. It was 0.4% down. And in terms of yield compression, we are talking about a yield expansion of 11 basis points. We look at your logistics, it was 0.6% down and seven basis points of yield expansion. And in shopping centers, it was 0.3% down and 14 basis points of yield expansion. This means that in the case of offices, we are not on a passing yield of 4.8, coming from 4.7. In logistics, we are flat, 5.6. And in shopping centers, we are 6.2, whereas we used to be 6.1. The big counterbalance of this negative valuation that was nevertheless very small in the traditional asset classes was data centers, where a 13.3% valuation increase in the first half. This is as a result of, as Ismael was already commenting, this asset class on which we have bet on a very interesting yield and cost, and it's becoming a reality. As it is becoming a reality, we are enjoying the valuation uplift attaching to it. Overall, we have a 5.2% passing yield for the whole portfolio, and with 12 basis points of yield expansion within the first half of the year. If we go now into the financial structure of the company, in the first half of the year, we have been able to keep our net debt pretty in line with the one we had by the year end. Loan-to-value of 35.6%. The average net rate is 4.8 years. And we keep our liquidity, which is even some debt that it used to be by the year end. These are 5.6 billion dollars. of liquidity within the company. The company has been active in the debt market in the first half of the year. As a matter of fact, first we were putting a top of 100 million on the bond starting in 2029. We have also been raising two mortgage loans the market. One with Caixa, 150 million euros. Another one with Novo Banco in Portugal. So we have been able to finance our assets for the first time in Portugal, putting financing into it with Portuguese banks. This is the first time ever we have done so far. The two of them are on a seven or ten year basis. So pretty compelling financing as well. That has resulted in maintenance of our net debt

and going to value.

Important to highlight that we got from S&P the triple B plus rating on a stable basis. We are just waiting for Moody's to provide us with annual review that we should be expecting either by the end of this month or early September. Not yet sure when they are going to be coming to with their annual review. If you look at page 25, you can see here that with the different financing that we have been putting in place in the first half of the year and two ones that we were putting in place in the last quarter of last year, we have been able to tackle the refinancing of the $600 million bond in May 2025.

So the company now has started to work on the refinancing of 2026.

which is November 2026 for the 800 million bond starting at the moment. So the company is very well set. We have almost all of our debt is hedge, 97.2% is already hedge. So we're not hoping to interest rate evolution. And 85.6% of our debt is unsecured. So we only have 14.4% of our debt on a mortgage basis. So I think this is it from the financial structure standpoint. The company is now on expansion, as we have been talking about on data centers. Thank you, Miguel. Well, moving into logistics, as commented, we have merged the best one and best two plans so that, because we are reaching an end, basically, on day one, and so that it is clear for the market what we are trying to achieve with the different locations and construction activities. In short term, basically we have 33,000 square meters fully pre-lit and the other 73 square meters are illegal under a grid ahead of terms format and will little by little be transformed into full pre-lets in most of the cases before starting construction. The other 57,000 square meters are still under commercialization with good prospects. All this product will be delivered during the second half of 2025, so between summer and Christmas 25, with a remaining investment of €91 million and expected stabilized GRI of €9.5 million. So the yield on cost, including the land, is 7.5%. If you calculate the yield on capex at present, and this is why we are rushing to put it or to move it from WIP into operation, it's in excess of 10%. It spans throughout five different locations, Cabanillas Part 2, where we will be putting into the market a 60,000 square meter shed, San Fernando 3, where we are going to be putting into the market 11,000 pre-lead and another 32,000 which are under commercialization. Lisbon, where we have 33,000 square meters. Sevilla is just one model, 2,000. And Valencia, where we are creating an inaugural shed in the new park of 25,000 square meters. Regarding the need and long term, we will still have 350,000 square meters in our belly, of which 106,000 are with agreed head of terms. Part of them, or most of them, are located in Lisbon, where we have reached an agreement which is phased over time with just one tenant for the full takeout of our logistic capacity in the Villa Franca de Xira industrial park. I guess it will be public in the coming months. except the land that we have reserved for data center use. And then there is another little module in Seville that will be also built and is under a grid hot. The rest is topping up our capacity in Cabanillas Park 2, Azuqueca 3, San Fernando 3, and the Valencia-Vetera Industrial Park. But, you know, commercializing all these mid- and long-term pipeline would eventually bring another close to 20 million of additional rents to the company. We have commented on many occasions that our intention was to bring logistics to a total income, the one visible in our P&L, because nobody sees, of course, the income coming from FILSA in Barcelona, given the accounting method. but on the visible income to go as close as possible, $120, $110, $120 million. That is the objective of the company. So logistics will little by little compete with shopping centers for becoming the second income source of the company. I know all of you are flashed by our data centers, but for the moment, Logistics and offices and shopping centers continue paying the bills in Merlin, so it's important that we pay attention to this. They look like little initiatives, but they are very important initiatives in our case in order to continue extracting value for the benefit of our shareholders. Moving into digital infrastructure plan, to the so-called mega plan. Well, do you know the snapshot of that plan? It consists of deploying 260 MW of IT capacity in Spain and Portugal. The phase one is already operating, although not fully equipped. I will comment the details about the deployment in terms of equipment. What is now on the table is the phase two, which entails 200 MW of new IT capacity development mainly in Lisbon and Bilbao Arasur, although there is a little repowering that we can achieve in the Barcelona PLSDF data center, you know, because we have received an extra, you know, injection of power from the distributor, from Endesa. Okay. Regarding phase one, well, basically in Barcelona, We are fully supplied in terms of electricity. Total maximum design of this data center is 16 megawatts pending the expansion that I just commented. We have received and installed 10 megawatts of equipment at present and we are pending to receive and install another 6 megawatts of equipment that will be ready by the end of first Q2025. We are 100% booked in this data center. And in Bilbao, it's basically 24 megawatts, of which we have received equipment and installed 10 megawatts, or will be installed by the end of year, sorry. And the remainder, up to 14 megawatts, will be installed up to second Q2025. In Madrid, Getafe, The electricity is supplied, but we are having problems now with the transport of that electricity. As you know, Madrid is not the favorite destination of investment for our central government. So we have six megawatts in place, and the remainder, we will go and find them in a different substation. So that will require a number of works, and we will obtain them on a phased way until first year 2026. That one is 70% booked, and we cannot continue commercializing because we have no full certainty on the moment of reception of all the electricity. Regarding the financials of this phase one, Well, we are running or we continue running approximately six months late, as commented on past occasions. We haven't been able to catch up yet. And we haven't rushed also because one of our clients is also late on the reception of their own supplies, particularly GPUs. So, you know, we wanted to have 42 megawatts installed by end of 24, and we will only finish the year with 26, and 44, two more, will be only achieved by second Q2025. Regarding CAPEX, accounted CAPEX has increased to close to 300 million as of first half. We have another 55 million that will happen before year end, and the rest has been moved into 25 and beyond. So basically, we have, in a way, saved a little bit of capex, 50 million or less, that has been moved into 25 and 26. The stabilized gross yield on cost remains the same. The net yield on cost remains the same. As commented, that is mainly dependent on the transfer price of the land, which was relatively cheap at the time. And the gross to net margin remains in the region of 70%. This is basically what we are doing in phase one. We expect to receive more than 80 million of rents when stabilized in 2028, with rents shining mainly in 2026 and 2027, because in 2024 and 2025, they will still be affected by deployment delays that will not affect the backlog of total value of contracts will remain the same. And we might also benefit from a number of bookings moving into full format listings during this period. Regarding phase two, the campus of Rivavellosa de Ebro, what we call Arasur in the north of Spain, is now set to host 118 megawatts of capacity in two new buildings. We have one building existing with 24, so we will be adding 94 megawatts of IT capacity, of which 48 megawatts will happen in building number two, and 46 megawatts will happen in building number one. The construction teams are in place. We are waiting for the construction license momentarily. So hopefully construction can start in this location by end of the year. Regarding Lisbon, we are good for 100 megawatts for now distributed in five buildings with 20 megawatts capacity each. It is true that after conversations with market and mainly driven by the fact that AI occupies less space than cloud and co-location, we are thinking about repowering each building to 36 megawatts that will result in 180 megawatts capacity just in this location. That is This will happen over time. I mean, it won't happen in one shot. But it is important, particularly because we already have electricity in order to feed that IT capacity, which is obviously very, very, very good news. So going one by one, on phase two, there is the six MW repowering in Barcelona, having received 15 MW extra utility power from Endesa. Many of you are... will probably question why 15 megawatts of utility do not give at least 10 megawatts of IT, the reason being that it is a separate feed with a separate electrical circuit. So, and, you know, there are a number of, you know, inefficiencies in the use of that electricity, and we have a lack of space at the rooftop for additional chillers. So, you know, we cannot use all the electricity of this new feed and converting it into IT capacity. But anyway, those six megawatts will be very much welcome because the current share is already built. Regarding Arasur, 94 megawatts, for which 140 megawatts of utility supply have been obtained. In the case of the first 70 megawatts for the first building, for the Arasur II building, Those will be supplied upon construction with no further infrastructure needed. They are already onsite. The other 70 MW still need some infrastructure works in terms of aerial lines and some infrastructure in the substation that will make the connection works only to be completed by 4Q2026, but this is no problem because Although both constructions will be overlapped, the people doing the structural works of Al-Asur II will then move into Al-Asur I. The finishing touches in Al-Asur I will only happen more or less simultaneously to the reception of the power at the end of 26. Al-Asur II will be finished before. Regarding Lisbon, we are projecting for the moment 100 megawatts of capacity in five buildings of 20. We have been granted 250 megawatts of power, 110 megawatts from EDP in medium tension, and the other 140 by Red Electric Nacional, by REN, in super high tension. So very interesting also the way the Portuguese government is reacting to the data center you know, wave very proactively and trying to help the different operators that are, you know, entering Portugal and taking positions there. In terms of the CAPEX plan, well, you have here a timeline, which is our present, our base case, which is basically to do the bulk of the works during 25, 26, and 27. Start receiving some rents in 28 and be stabilized in terms of GRI in 2029. With very similar parameters, 4.2, stabilized GRI, more than 10% net and around 70% gross to net margin. This is our current BP and this is the assumption with which we are working at present based on conversations with the counterparty, with the clients. In case we go little by little because we only have partial bookings, then eventually we will develop only building 02 in Arasur and two buildings in Lisbon. So that will mean between 48... in Arasur and 40 to 72 in Lisbon. So total power will be between 88 and 120 more or less. So this is what we will start doing in case we cannot do an agreement for a full takeout of our existing data campuses. On page 37, We simply wanted to introduce the fact that there is more capacity in those two sites only. I mean, without talking about other eventual pipeline locations. In just those two sites, there is extra land. We have been buying a number of plots in Arasur in order to make sure that we have extra land for a full development off the data campus with, you know, what the Americans call disability 300. So, of course, we have full disability up to 300 megawatts. More than 80% of the electricity required has already been granted, and we are waiting for the rest, you know, and we'll try to obtain it during the coming months. In Lisbon, we have reserved For sure, one additional plot of land, which is plot number 23, and contingently we also have another plot of land, which is ER marked for reserve in case we need it in the future. Again, with the idea of having visibility 300 in case of need. Power for these will be requested. Of course, I guess it will be obtained in the long term, but it will be requested in due time. On page 38, you can see the exact location of the different plots that will give rise to the extension of the Arasur data center campus. Out of the approximately 450 megawatts of power that would be needed for 300 IT, as commented, around 375, 80% is already granted, and the rest is under request. And in the case of Lisbon, leaving aside the 80 megawatts potential repowering of the existing five buildings, we have planned for an extra 120 megawatts IT in an adjacent plot for which around 200 megawatts of utility power need to be requested from Red Electrica Nacional, which will be done momentarily. In order to sum up and finish, very strong performance in all key financial and operating metrics, whether like for like rental growth, occupancy, risk spread, or FFO generation. Very healthy occupancy levels in all three asset classes, traditional asset classes, with the offices performing very, very solidly. And of course, I know it is a little bit strange in the current world that For some reason, the market in Spain, as commented on many occasions, remains in equilibrium. There is no big problem of large supply, at least in Madrid and Lisbon, so very good performance. And virtually fully occupied in logistics. We will end up the year around 99, and shopping centers, we will end up the year around 96. The FFO guidance for 2024 is confirmed at 0.59 per share. I know that many of you are already wondering why don't we raise the guidance. A guidance is a guidance and we try to be prudent. We know for sure that the second part of the year is going to be heavier in terms of overheads. Why? Because we have now set up commercial offices for our data center division in the Netherlands and the UK. because all the equipment that we are installing needs to be maintained. So that affects, of course, our FFO. So we have drawn on a number of trade lines recently that we, you know, create more financial expenses that we'll drag also on FFO. So we prefer to be prudent towards year-end, and we are just confirming our guidance of 0.59, and then only God knows what will happen at the end of the year. In terms of maturities, no problems until November 26th, and our intention is to recommend to the board of directors a total dividend of 0.44 for the fiscal year 2024, which, of course, should entail a payment before year-end on account, as we have done on many occasions, and the rest will be paid next year following the general shareholders' meeting approval of the full-year accounts. That is all. Let's move into Q&A. And let's see. I expect a lot of questions about the capital increase, but you are all capital market professionals. So please. Okay?

Thank you, Ismael. So we remind you that for those who want to raise questions, please press star followed by number five. We already have five. people on the line. The first question comes from the line of Jonathan Conato from Goldman Sachs. Jonathan, the line is yours.

Good afternoon. Thank you for taking my question. First question, I wanted to talk about the timing of the data center facing, in particular, Phase 1. Also, thank you for having provided the capex on Phase 2. Is the timing on Phase 1 aligned to what you had said earlier, like Q1? I mean, obviously, we've talked about six-month delays. Now you're talking about CapEx that is a bit more spread around to 26 and 27. So just trying to understand if there is any change here and also how are your discussions progressing with customers and this, effectively, deployment and the feeling of their own capacity. Is that on time or is that just taking time because it's complicated to get equipment? That'll be my first question, please. Thank you.

Okay. Look, Jonathan, Clearly, there is a mix of situations regarding the delay of phase one. On one side, it is true that our clients are now finding increased difficulty in getting their supplies, particularly TensorFlow GPUs. So in one particular case, they have already told us that they are going to be delayed on the deployment by about six months. This is the bad news, because that will have an effect on the year-end cash flow, even though it's going to be minimal. And in 2025 cash flow, because we were counting on having the cash flow from the very beginning of the year, and it will only happen in the second quarter. However, the good news is that the backlog remains the same. Length of contract and escalations, everything remains the same. And eventually, we might move the booking into a full-format listing towards the end. So that will be also very important for us because it will provide a lot of visibility to the market regarding that income. In Europe, as commented on many occasions, so far the problems are minimal as compared to the U.S., talking about MEP equipment, the equipment that pertains to us, data center operators, not talking about GPUs, because GPUs, there's virtually one monopoly of just one maker other than the GPUs that Google is making for themselves, but the rest are normally bought from NVIDIA. In terms of MEP, in the U.S., there is clearly a problem In Europe, we are starting to see some hints of problems. We have been late in receiving a number of generators and a number of skids that should have arrived at the end of June, beginning of July, and will now only be received in October. So this is happening. I know for some people this could be a big problem. I believe this is part of life, happens all the time in development. So we take it with a little bit of cold blood and simply live with it.

It's a developing area. Just following up on that, do you have any penalties on any of the sort of delays that can happen or is it just that you're receiving the income late?

You have or you might have penalties but you have to be very careful. So Probably better than the penalties, what you have is alternative sources of income. I mean, you might decide not to wait for a certain booking and make use of your other booking in order to bring an alternative source of income that you need to weigh the quality of the client because not everybody in this industry is the same credit worth.

Okay, and generally speaking, like the interest from clients and the bookings, I mean, obviously you're saying, I think that pre-bookings for a phase one is something like, I think 80%, if I remember correctly, and sorry, 90%. So that's quite high. So the interest from clients is developing according to your plans. And is there any time where you convert to leases already? How many are converted to leases and any clients you're able to talk about already?

Yes. Well, the amount of bookings and interest in the market is clearly surprising us on the upside. New companies are appearing every day. Those are, in most cases, beefed up by private equity money. So in many cases, they are, in a way, well capitalized or willing to take significant blocks of power. But you have to be a little bit, particularly because This is an activity in which we plan to stay forever, so you have to be a little bit selective in what you get because you want an activity that lasts forever. You want to occupy many other blocks of IT power, so you have to be careful with that. I mean, it's like in logistics. We, of course, tend to favor existing clients and people we have a relationship with. So this is the situation. Another thing that is surprising us is evolution of rent. I mean, clearly, rents are subject to certain tension. And nothing, I mean, compared to what we originally underwrote is radically different. But anyway, I mean, we don't want to be too positive about it until the cat is on the back. I mean, let's wait. And then regarding moving bookings into leases, the biggest obstacle remains, as commented on many occasions, financial legislation. But yes, I believe, generally speaking, you will see most of those bookings moving into full-format leases during the year because it is in the best interest of both parties because people want to reserve IT power. And in our case, we want certainty of income, which, given the overbooking, it's certain that you prefer to have it in writing in a full format contract.

Okay, very clear. One final question on timing. How long can you go effectively given your plans without raising the equity?

For the phase two, well, the maximum till year end, but it's a chicken and the egg situation. I mean, of course, you want to wait as much as possible in order to avoid, let's say, the cash drag. Call it that way. I mean, having money on the bank account. But on the other side, you want certainty of funding as soon as possible because your clients are also looking to you. I mean, they want to make sure that you are able to deliver in full what you are negotiating with them. So, you know, they don't want to... Yes, they know you are a listed company. They know you can do many things. But, you know, the sooner the better. They want to see, of course, certainty of funding so that you can do what you need to do.

Okay. All right. So the next question comes from the line of Logan Laos from Odo. Logan, the line is yours. Thank you.

Yes, good afternoon. Thank you to take my questions. So maybe I would have first a question on the valuation of the assets. So first on the current operating classes, we have seen that offices, logistics, and shopping centers are quite stable, and we see a strong increase in data centers. So is it possible to have maybe more colors on how it has been valued? And is there any potential still uplift to come on that portfolio? Then maybe on data centers for phase one, I would have a question. So would it be possible at the end, due to the delay, to have maybe more visibility on what do you expect in terms of revenue in 2024 and in 2025? for data centers. And then maybe on phase two for the funding, so we understand that maybe you are working on different options. So would it be possible maybe to have an overview of these different options that you could develop depending on the interest that you can find from third-party investors?

Well, regarding the appraisals of the different asset classes, we have recently had our audit committee and the appraisers came and explained their evaluation methodologies. In the traditional asset classes, you all know how those assets are valued in offices. All that we can say is that we saw more pressure, more downward pressure on the values in Barcelona, a little bit more peaceful regard to Madrid, and of course, a very bullish with regard to Lisbon. In shopping centers, they are starting clearly to stabilize. I mean, they are starting to like them. You know, the trend has clearly reversed while they continue adjusting because, you know, there is transactional evidence that in principle yields are higher, et cetera. While they continue adjusting, but in shopping centers, regarding operational metrics, it is clear that they like them a lot. On logistics, the tone is flat on the existing assets. I think the yield expansion is probably ended in logistics. And in terms of WIP, very positive. I mean, we have already experienced a significant uplift on the width that we have moved into operation. And we'll continue to have the same effect as we bring more and more product into operation over the coming years. I mean, you have seen our numbers. I mean, around 200,000 square meters are already under head of terms. So as they come into operation, we will enjoy a significant valuation uplift. Regarding data centers, I was curious, as you are, about how they value them. I saw that they discount cash flows for the next seven years, and they only assume stabilized cash flows in 2028. I'm talking about phase one. The exit yields that they assume in year seven, I believe they are conservative, between six and a half and 675. I think it's okay, talking about net yields. The GRI, their projections grow GRI by around 2%. So nothing galactic. I think everything is more or less reasonable in my view. Then the discount rate is very high, in my opinion, but I will not disclose it. And, you know, that is the reason why we, little by little, increased the value recognition of this asset class, which at present has only meant around $120 million of added value in our balance sheet. But just for the 60 million of phase one, we are expecting between $500 million and $600 million net impact in terms of value creation just from those 60 megawatts of capacity. That is basically the way they value phase one for which we have little intervention, if any. I mean, they value us existing assets, okay? So the only point of contention is basically the discount rate. Then regarding the 200 megawatts of powered land for phase two and further expansion capacity, this is the only part of the equation in which we have a say. And our say basically is no value. So the existing land is to be valued at cost. We will continue incorporating capex, and we will basically start valuing if and when inaugurated. So little by little, the value recognition will take place really on a staggered basis. And then for further expansion capacity, which of course will be very attractive to bring forward the valuation and fuel a little bit share price, et cetera, our instructions is it is out of scope. And therefore, it is carried at historical cost, which is very, very, very, very minimal. And that is the way we prefer to keep it. We have received also a lot of questions from some of you, some particularly analysts, regarding why don't we capitalize financial expenses interest on developments. We prefer to expense them. We prefer to do it that way. And we believe this is a fairer picture of where we are in the company. Of course, there are many schools of thought regarding this, but up until we are not obliged by either our auditor or the accounting authorities, as we were in the case of some examples of linearization of expenses, the vast majority will continue not to linearize, but we were obliged in long contracts, we were obliged to linearize some expenses by the auditor. But other than that, until we are all right, we prefer to simply expense as they happen. We believe it is more prudent to do it that way. Then you wanted color on the funding options as commented. I think it will be to no avail at present to enter into what funding options we have. shouldn't speculate. The board at some point will be convened and will take decisions and we will simply execute. So stay tuned. And regarding income for 24 and 25 later during the year we will provide more accurate numbers. For the moment we only know that of course our income in 2024 is going to be significantly hit as a consequence. Anyway it was an irrelevant income. Don't be afraid. but it's going to be hit by the delay in the deployment and part of that impact will also be felt in 2025. Nothing in 26, 27, 28, 29, but of course in 24 and 25 we will have an impact and we will quantify towards the end depending on how things evolve regarding the recognition of that income and eventual upfront payments. Okay?

Okay. Thank you very much. That's very clear. Thank you. Thank you.

Thank you, Florian. The next question comes from the line of Celine from Barclay. Celine, the floor is yours.

Hello, Ismael. I just have one question very quickly on your OPEX non-overhead. What is driving the 5 billion increase compared to last year? And how do you expect us to model this going forward? And also you mentioned something around UK Dutch offices. What is this for?

Okay. Well, UK Dutch is basically commercialization offices for data centers. You know, the decision making of most of our clients is located in those two regions. So we need to have a permanent sales representative position. in those regions, so that, of course, has an impact in expenses. Okay, so we now have personnel, employees, working in those locations. And regarding OPEX, Miguel? In terms of OPEX, it's clear that we have nine megas in operation for the first half of the year, so that has an impact with low income so far because of how we are already evolving in commercialization.

Sorry, you meant the non-overheads.

Yes, the OPEX non-overheads.

Sorry about that. I didn't follow the question. Now, this is a one-off, and it's related to the financing we have been putting in place in the first half of the year. As you know, we have been putting in place two mortgage financing, one in Portugal, one in Spain, and this has a of course, attached to the mortgages. So we're talking about 4.4 million euros which are attached to this specific financing that is a one-off. That's the additional behind the increase in non-order heads with regards to the periodia, whereas the non-order heads were more or less recurrent, more attached to the purely rating and things like that.

So that's a one-off?

Yes, sure.

All right. Thank you. Thank you, Celine. The next question comes from the line of Ignacio Dominguez from JD Capital. Ignacio, the line is yours.

Yes, good afternoon. Thank you for the presentation and taking our questions. Just one from my side. This one is on asset valuations. Do you expect mortgage expansion in the second half? How do you see the market in terms of number of transactions? Any improvement in liquidity? It would be nice if you could detail your views by key operating segment. Thank you.

OK. Well, the ones that I picked, yield expansion, I have proven to be a complete disaster as a wizard regarding yield expansion, because my crystal ball is completely broken. I told you that I expected offices to move beyond the 5% you know, during the year, and we are still at 4.8, and frankly speaking, I do not expect that they will move beyond 5 as of year-end, but if they move, I wouldn't be surprised. I mean, I am of the opinion, and I am very frank about it, that offices need to move to around 5, 25, 25 easily because the market wants that. It is like shopping centers. The market wants that, so, I mean, there is nothing... in trying to deny that this is a market trend. So I believe there should be more expansion in offices. However, in shopping centers, the tone I have seen from the appraisers is now more positive. Let's see if they continue adjusting. But the tone I've commented is more positive, so I wouldn't expect more significant corrections in value other than maybe not translating the like for like into value, which is already devaluation. Not moving the like for like increase in income into value obviously means a devaluation of the asset. And in logistics, I also expect a relatively muted or flat trend towards year end. But as commented, I could be completely wrong. Regarding liquidity in the different sub-markets, clearly the liquidity in shopping centers is increasing, both for smaller transactions and for bigger transactions, although, of course, nothing really, really big has taken place so far. Lots of rumors that nothing has yet happened or transparent to the market, other than, of course, the capital markets transaction of Grupo Lar. In offices, we continue seeing reasonable liquidity on the small tickets. The protagonists are, of course, the family offices and no liquidity whatsoever in the medium and big ticket sizes because, of course, institutional investors are not into offices at present. In logistics, the activity is relatively limited. that is mainly because there is no product available in the market and there is not a lot to sell and therefore there is not a lot to buy in logistics. And then in data centers, I mean, if you were to put one in the market, everybody would like to buy one. But this is not an intention to put them in the market.

Okay, thank you very much. It's a pleasure.

Thank you, Ignacio. So the next question comes from the line of Mark Mossi from Bank of America. Mark, the floor is yours.

Thank you. Very good afternoon, all. I have a question around your guidance, FFO guidance. Should we just consider that you're very conservative because you're implicitly assuming that you're going to create new shares at a time you're going to raise equity?

No, look, our guidance, I mean, we are talking as a present, As we are talking today, we are talking as in macroeconomy. Therefore, we are not making any assumption on anything, on new equity, on financial income of the new equity. We are not assuming anything. We are just, in a way, trying to project what will happen in the company if it remains as it is at present. And the reason why we don't increase guidance and we remain prudent, as commented, is mainly because we expect more significant overhead during the second half because we continue equipping our data centers. We continue hiring people. I think the total staff in the data center division, I think, is now 27, 26. And it's set to increase to around 32 people by year end. So it is a division which is growing. Many of the persons we are adding to that division are, of course, senior professionals, well-paid. I mean, of course, if we want to continue operating data centers for many, many years to come, of course, we need to have a credible team like we have in all other asset classes. And, you know, We are taking that into account into our second half projection. The second reason is we have an increased financial cost because we have tapped a bond and we have entered a couple of new loans in the first half, which have not affected the first half or at least have not affected the first half in full, but will certainly affect the second half. Yes, we have cash, but the remuneration of that cash is also going down at present. So the delta between what you pay for a loan and what you get for money at banks has increased versus other situations in the past. Therefore, we prefer to be prudent. Do we expect to beat it? Yes, but we cannot say anything further. I think it's better. There is no gain for us or there is nothing good for us in being too bullish about how are we going to end up the year. We prefer to to see what happens during the third and the fourth quarter. Remember, Mark, that in the last month, in terms of clear market, we have had the first round of the French elections where the fear was Le Pen, the second round of the first French elections where the fear or the reality is Mélenchon, the attempt of assassination of Trump, And yesterday, you know, the resignation of Biden. You know, it's hard to find one day of rest in this brave new world. So we prefer to be prudent.

Yeah, I understand that. Can I ask the question differently? What sort of yield do you get on your cash in your bank right now?

The best thing we are getting at present is like 3%. 30, 50, around that is the best remuneration we are getting. I mean, if you know one of your cousins at Bank of America that gives us more than four, we are happy to make a movement.

I have no idea. Thank you very much. Thank you very much. But it's interesting because when you do the math, meaning you end up at 0.59, even in creating new shares for half a year. So that was just the purpose of my question, even with 3.5% or 3.25%. My second question is about the dividend for the first half. Why did you not disclose it right now?

The dividend? No, the dividend, we are disclosing our recommendation to the board is going to be 0.44, which is normally made up of 20 in October or so. I mean, last year we paid it a little later, but normally October, November, we pay that dividend, which is a dividend on account. And then the remainder is normally paid after the general shareholders meeting approving the annual accounts. So the rest of the dividend will be paid next year after the general shareholders meeting.

What I mean by that is we should assume 20 cents paid in H2.

Ah, yeah, but we need to go through the board of directors. I mean, it is not within our powers to distribute the dividend. I mean, to recommend, yes, but not to distribute unless we are instructed by the BOD, and the BOD normally takes place around October for the approval of the dividend on account, okay?

Okay, but traditionally you never announce the proposition of a half your dividend.

No, we normally pay a little less than half. Again, out of prudency. I mean, there will be nothing wrong in paying 22, but we normally pay something more like 40% instead of 50, and the other 60 we pay after the general shareholders meeting.

Okay, that's clear. My final question is around your 180 additional megawatt capacity you're foreseeing on your land bank. What sort of capex should we assume? It's still 10 million euros per megawatt. And what sort of timing should we start to eventually consider here?

Yeah, this is an exercise that we have tried to make It is impossible at present to determine the capex because just by the weight of things, just developing the next 200 megawatts of capacity is going to keep us busy till at the very least. I mean, regarding the potentiality of thinking about a farther phase, we are going to be busy till easily the end of 26, beginning of 27. So only then we could start thinking about developing further capacity. And basically, we don't know what is going to be the cost of prefab concrete, steel equipment, particularly equipment at the time. We don't know what the prevailing rents are going to be. So it is a little bit premature. And also, frankly speaking, we don't want to clog the market. We don't want the market to kind of feel overwhelmed. about our capacity. What we have is simply a value reservoir. It's basically a value storage, which is there, and it's going to be positive at some point in the future, particularly having received the power. In the case of the Basque Country, we are relatively close from getting all the power we need in order to bring the data center campus to the 300 visibility, as clients normally call it. And in the case of Lisbon, it's a little bit more binary because we are depending on the outcome of an additional utility request that we have to make from the Portuguese authorities. But having seen how they have responded in the past, we are positive. So this is what we are. We wanted to disclose that there is further capacity just as a way to explain to market that the music doesn't stop at 260, that it's very, very early to make any calculations about what is going to be the total cost, et cetera. But, well, as commented particularly with you, Mark, in some occasions, of course, our long-term ambition is to be a relevant company in the field of DC operation in Europe. Of course, there will be many others, but we want to be one of them. We want to be in that league.

Is the ideal breakdown between data center logistic and the rest? For you?

Look, if with 260, the logistics and data centers already will represent around 60% of the income. If you add the extra income from the rest of the power, you know, clearly the company will go north of 80% eventually in terms of what we call digital income.

Brilliant. Thank you very much, Ismael. Have a good rest of your day. Thank you. It's a pleasure.

So the next question comes from the line of Colm Marley from Qualytics. Colm, the floor is yours, please.

Hello, thank you for taking my question. Just three quick questions, please. You spoke about beginning how the increase in indexation has reduced some of your releasing spread in the office portfolio. How are you thinking about balancing occupancy levels and rental levels going forward there? A second one of data centers, I haven't noted that in the Madrid data center bookings are only at 70%, but the others just wanted to clarify if there was a technical reason for that. And then finally, you mentioned earlier that you had experienced some delays in receiving generators and other machinery, which are pretty critical infrastructure to your data centers. Just wondering, are you currently thinking or having any discussions on how you could minimize or hedge these risks if geopolitics or supply chains get worse?

Very good question. Look, regarding the delays, it is true that what we have seen so far has been still relatively minimal, affecting some skids and some generators. The construction times remain between 15 and 18 months for the moment. The way to minimize is basically to down pay and be less speculative when ordering, but that results also in the anticipation of significant capex. So we normally are careful, or have been in the past, very careful about doing that because around 70% of the cost of one given piece of equipment is normally paid after the testing of that equipment that takes place upon installation between 15 and 30 days following installation. You test it and you pay, and that is around 70% of the money. The other 30 was normally downpaid before. So is there a financial way to hedge These, I don't know. Maybe there is one, but at present I cannot think of one. The purchases are made in euros. That is important. Even in the case of the American equipment, we are paying euros for the moment. So, you know, there is no FX hedging involved for the moment. Of course, you know, we are subject to the valuation in dollar terms of that equipment compared to Europe, but that's it. So that is all I can say. I mean, for the moment, it is not like in the U.S. where, you know, if you want to get... supply chain disruptions, etc. Well, the supply chain disruption that mainly worries us will be U.S.-Europe merchandise flow because the rest of our equipment comes from OECD countries, namely Spain, France, Germany, Sweden, U.K., some Nordics, and that is basically it. So, in principle, I mean, the supplies coming from countries which, in theory, could send you the equipment in a truck rather than in a boat in case of need. We are not subject to problems with Asian or Chinese equipment. We don't equip Chinese. So, that is basically my take on potential delays? The range of what? In the traditional asset classes, you want my comment on where the occupancy could go from here? Very difficultly, up. I mean, in logistics, going above 99 is going to be complicated. in shopping centers, given the high rotation of tenants, going above 96 is very, very difficult, if not impossible. 96 is already full occupancy in shopping centers because there is always rotation of tenants. It's subject to significant attrition. Many concepts evolve and die, and you need to replace by others, and that normally keeps always a certain structural vacancy. The only asset class in which we could add a couple of points of occupancy could be offices. But that is basically it. And the effect in in cash flow will be limited. That is for the let's say upside risk for the downside risk. Of course, my fear is with macro economy. I mean, we have proven very, very resilient to, let's say, fashion waves. I mean, shopping centers have resisted the fashion opinion wave of the e-commerce, and offices will resist the fashion wave of the work from home and all that. However, they remain human-related assets. So if the economy starts going down, of course, if there is unemployment, there will be destruction of office space, and as such, occupancy will increase. Sorry, vacancy will increase. And in shopping centers, if for some reason there is less consumption capacity in the Spanish households, of course, or a decrease in tourism, which is also helping, of course, the shopping center will start selling less per square meter at the beginning. And then as a consequence, either vacancy will increase or rents will start falling. Likewise, in the case of logistics, which depend heavily on e-commerce. So it depends on the evolution of e-commerce. For the moment, it's clearly positive. No longer double digit. Now we are talking about single digit, but the evolution continues to be positive. So we continue to enjoy positive momentum and rental tension in logistics. As commented in other occasions, CPI indexation and release spread are enemies, one of each other, because as you pass on inflation to the tenant, you are obviously taking a vent on your reversionary potential. So at some point, of course, you catch up with market trends. And once you catch up with market trends, what you can expect is basically a flat trend. And at some point, if the economy starts to suffer, of course, it's going to be a negative trend. That has been like that, you know, for ages. I mean, our activity is subject to cycles, and there is nothing about complaining for it. I mean, simply when the cycle is not good, you suffer a little bit on vacancy and you suffer a little bit on rent, and that's basically it. But this is the reason why we also are a relatively diversified company, so that we can play the cycles insisting more in one asset class rather than another asset class as we did during the COVID, etc. So we are accustomed to that. And then overall, it is also important to say that overall rent in Spain, believe it or not, remained relatively cheap. I mean, at least remained relatively far from the peaks experienced pre-great financial depression. So while Prime rents, prime average rents in CBD in Madrid in offices can be at present 36, 36 and a half euro. The top rents achieved prior to the Great Financial Depression were in the region of 44, 48 in some cases. In logistics we can tell you that for example in Barcelona we charged or the park was charging rent to clients of close to 10 euro and at present we are at 7, 25, 8. in the best cases. So there's still some room compared to the highest historical rent. And in shopping centers, what can I say? I mean, in shopping centers, clearly the rents are lower than they were, but also so is the occupancy cost ratio, which, as you know, in the countries in which there have been big problems in shopping centers, normally those problems have happened when OCRs have gone above 20%. So this is what I can say. Of course, I remain prudent because nobody knows what the future might hold. This is the reason why we decided to launch our activity in data centers. In order to have a clear avenue of growth in the coming years, in anticipation of the fact that we knew at some point we would be reaching such a level of optimization in our traditional asset classes that there would be very limited upside.

Thank you. It's a pleasure.

Thank you, Colin. So the next question comes from the line of Fernando . Fernando, the floor is yours.

Hi. Thank you. Thank you for taking my questions. I will make three, please. First, with regards to the pre-bookings, last quarter you quantified that you were in early conversation for a 30 megawatts pre-booking. I don't know if you could update on this amount and also on the conversations you're already having, what sort of amount or tickets of megawatts are you discussing right now? Are we talking about 20, 30, 40 megawatts per client or smaller, bigger amounts? Second question with regards the OPEX leakage in for phase two. So I've seen that you lower the leakage below 30%. So I don't know if you can comment on the reasons behind this improvement. Also, I don't know if you can comment why your business model is way more efficient than that of, for instance, digital reality, because your margins, net margins are well above digital realities. And last question, you just mentioned that the portfolio is reaching, let's say, a mature level at some point, offices, logistics, and shopping centers. Will you consider selling a big chunk of assets in any of these categories?

Okay. Let's start by the end, maturity of the portfolio and possibility of sale. As we have commented on many occasions, we are not Apex partners. So we don't buy on the trough and sell on the peak because that is, first, it is very difficult, and second, there is no liquidity at present, so there's nothing you can sell. And if you sell in bulk numbers, if you were to sell 6 billion worth of offices because you think they are mature, imagine what the buyer thinks. And the buyer is going to take that into account into the pricing. So, yes, we continue rotating mature assets. We continue... using or making the best out of the different trends that we see in the market. At present, we are clearly surfing the trend of residential reconversion. So we continue selling some of our non-core stuff. We continue refining the quality of our portfolio, which is something that very few people really have picked up in recent years. I mean, the quality of the portfolio of the company, which once was highly criticized by everybody, et cetera, has, I believe, significantly improved. And you can look at the presentation, which is now hanging on the web. And if you look at the office buildings that are pictured in that presentation, that is 86% of the value of the company in offices. no longer a mix, as many people say, a mix of offices. I mean, there is a big difference from our initial stages of development of the company in 2014, 15, and we have made sense of our business plan. Clearly, we had a business plan. We put together a big company, and then we started refining, digesting, and regurgitating assets and Clearly, we have improved significantly the quality of our portfolio. There is no, at present, no deferred maintenance. Our buildings are completely up to date. We are refurbishing everything in which we can see that we are going to capture value potential. So we are refurbishing offices as we speak. We are developing Liberdade in Lisbon, which is going to be a landmark asset that is going to be probably the best asset in Portugal. So we are doing lots of things in order to make sure that our portfolio continues yielding. I know that the immediate reflex of everybody is sell it, and then buy something cheap, and then make it expensive, and then sell it again. The regime is also not very good for that because every time you do one of these things, then you have to pay special dividend, you have to pay deferred tax losses, you have to pay the municipal taxes, you have to equalize the debt. So at the end, the money you can recycle from selling assets is very, very little. I mean, we have made this explicitly, this exercise, for the benefit of our board of directors because, of course, there was that, I would say, opinion that sell all the logistics and then put the $2 billion in data centers. Okay, if we sell $2 billion of logistics, we are going to put in data centers $300 million. So, you know, this is important to make that reasoning because, you know, sometimes people get a little bit obsessed about doing things the Disney way, which is not the way things should be done in an industrial company like us. Then regarding the OPEX at slightly better than 30%, this is simply because we are little by little incorporating some operating improvements, i.e. things that we have observed that we can either procure from a local provider. There are refinements that we can make in the shift. There are a number of things that you can do in order to try to slightly, very slightly improve your operating margins, which is something that of course worries us because we come from traditional asset classes in which the gross to net has, I mean, if you have a gross to net of 30% in logistics offices or shopping centers, you are clearly not operating But in data centers, it is heritage from the past, and the client pays you rent net, and they do not take care of operating expenses. And as such, we need to break our brains in order to make sure that we can, little by little in the future, reduce that leakage between gross and net. I am sure that just by sheer pressure of rent, at some point, it will not be increasing rent, what you will be discussing with clients. It will be a function of expenses. So at some point, that could improve a little bit the margins. Why margins in Spain, not ours? Why margins in Spain are better than in the U.S., not digital realty? Why margins are better in Spain than in the U.S.? Cost of labor. I mean, simply as simple as that. an average engineer or MEP maintenance worker in the U.S., it's easily three times what it is in Spain. And as such, this is something that we have exchanged information about with our cousins of Koch Real Estate in the U.S., and it's very similar here than it is there. Regarding pre bookings, the 30 for Lisbon was a pre booking in case we were to develop their in, let's say, on a staggered basis. So if we were to develop a first building of 36, yes, we have conversations for 30 of those 36. Normally, we would develop two buildings, so it will be 72 of which 30. It's in conversation. The reason why we have modeled food construction both in the Basque Country and Lisbon is because it is in line with what we can see at present. Only God knows what is going to be the outcome of those conversations and statements of qualification that if we can, of course, we would love to develop and lift up in full those two big schemes, because the risking exercise that will bring to our books will be extraordinary and clearly will you would place us in a different planet. So you know, for the moment, this is the most I can comment about status of discussions regarding the phase two.

Okay, thank you very much.

Welcome, Fernando.

Thank you, Fernando. So there are no further questions. We thank you all for being with us and joining this call. It's been a long one. We hope that all your questions have been answered. If not, you know that we always remain at your disposal. So please call us or send us an email. Thank you very much. Have a nice day.