Tritax Big Box REIT plc (BBOX) Earnings Call Transcript & Summary

Well, this is unexpected, isn't it? Here I am walking along a golden shoreline, soaking up the sun, a world away from the hustle and bustle of the square mile. You might be wondering, can this really be the setting for today's investor event? Has the CEO of Tritax Big Box taken the day off. But as we all know, appearances can be deceiving. What if I told you this scene is all thanks to advances in modern technology. This setting, not real; the sunshine, not real. Me, my voice, my favorite holiday shirt, all cloned, if you can believe it, and all thanks to the incredible advancements in generative AI. Imagine that just 2 years ago, this level of realism would have been unthinkable. Yet here we are in a world where cloud computing and AI are reshaping what's possible. These rapid advances aren't just transforming entertainment. They're changing the very fabric of industries and data centers are the backbone of this revolution. They power everything from AI breakthroughs to the cloud infrastructure that enables seamless global scale technology. But this pace of innovation is exactly why Tritax Big Box is complementing its years of experience and expertise at the cutting edge of logistics with capturing the extraordinary opportunities in the data center space. We believe we're only in the foothills of this technological revolution. And just like we were over a decade ago with logistics, we are applying our innovative and agile mindset to this exciting and adjacent real estate subsector. Thank you for joining us as we dive into these 2 growth stories. Today, we will explore how Tritax Big Box is uniquely positioned to deliver superior risk-adjusted returns with the potential to grow adjusted earnings by 50% by the end of 2030. But before we get underway, there's one thing AI hasn't mastered just yet, an AI-driven CEO. Thankfully, we're still some way off from that. So I'll stay here enjoying the sunshine. In the meantime, real Colin, over to you. I think you did rather better than me actually. Well, welcome, everyone, and thank you so much for joining us. Attractive as that beach looks for my digital doppelganger, I'm very pleased to be here in person today to share with you in more detail how we're powering earnings growth across our business and why we see a very positive outlook based on our clear strategy and our strong track record of delivery. So turning to this afternoon's agenda. I'll kick off with the key messages that we'd like to take away from today before handing over to Henry Stratton and Andrew Dickman to talk through the superior returns that we're delivering from our logistics development opportunities. As Head of Research, Henry is our expert on the market, and Andrew heads up our development activities. We'll then explore our data center pipeline in more detail, and I'm very pleased to welcome Kevin Restivo from CBRE, who is going to share his insight on the data center market. I'm also pleased to introduce to you Tim O'Reilly, our Director of Strategic Power, who will set out how we have secured the key raw ingredients for our data center strategy. Then Frankie will explain what this means for our earnings growth and how we're delivering superior risk-adjusted returns. At the end of each section, there will be an opportunity to ask questions. And the more interactive that we can make this, the better, please. The one thing I need to emphasize is the advice that we are heeding from our advisers and legal counsel. And that is that we are restricted on what we can say about the recommended offer that we announced last week for warehouse REIT. But I can say this, the proposed acquisition aligns well with our strategy. It complements the first of our 3 growth drivers: Active asset management to capture rental reversion and it reinforces Tritax Big Box's position as the leading logistics REIT. Obviously, there is still some road ahead, but the transaction, which we're very excited about, will provide a combination that would enable us to unlock the full potential of warehouse REIT to drive returns for all shareholders. And one final bit of housekeeping for this afternoon. Unfortunately, given that we're in an offer period, the home team will have to take a rain check on the social function that we organized to follow today's seminar. But the good news is that you, our guests, are welcome to enjoy canopies and refreshments on the terrace after this session. Naturally, we are very disappointed not to be able to share a drink with you all, but we thank you for your understanding, and we hope that the exciting elements of what we are showcasing today will go some way to make up for it. So in terms of today's main takeaway, the key message that we want to leave you with is that we can deliver superior risk-adjusted returns through our asset management, development and data center opportunities with the potential that by the end of 2030, we can grow adjusted earnings by 50%. And as we'll highlight, this potential is the result of our clear strategic choices. We've taken steps across all our activities to minimize risk and maximize returns to create a truly compelling offer for investors. Our strategy is exclusively focused on U.K. logistics. And we believe that this focus drives enhanced returns to shareholders, and we constantly seek ways to optimize performance and drive those returns further. You can see at the top here that ever since our IPO back in 2013, we've consistently looked at ways to optimize our performance, adding development exposure at the right time and more recently, complementing our big boxes with smaller urban and last mile assets. And as we announced in January, securing significant data center opportunities. And as you can see at the bottom of this slide, we've delivered attractive EPS progression. We've grown dividend streams, and we've delivered sector-leading total shareholder return performance. But the good news is that we're only at the beginning of our journey, and I've never felt more confident about the future prospects of our business. Key to the performance that we've delivered and to our future performance is the investment that we've continued to make in the team at Tritax management. And you can see on this slide here that we've got a fantastic team of experts in their respective fields, and we continue to invest in both the number of employees and their ongoing training and development. And this has been delivered incredibly cost effectively to Tritax Big Box shareholders with an EPRA cost ratio of 12.6%, which is one of the lowest in the sector and the lowest for a U.K. listed REIT with an active development platform. Ultimately, we believe that this is a highly effective platform to deliver outstanding performance and returns for shareholders. Now let's turn for a moment to our strategic choices. And as you can see, our model is very simple. We've deliberately chosen to construct a portfolio of high-quality assets that attract world-renowned customers. These assets commonly form the backbone of their U.K. supply chains. And this provides a strong income foundation, which we know for many of our shareholders is especially important. And from this strong foundation, we can enhance our returns through our direct and active asset management, including managing smaller assets. We also add an element of development to replenish our portfolio at very attractive yields. And all the while, we objectively review the assets that we've got within our portfolio, seeking ways to maximize value by rotating capital where appropriate into our growth areas. And it's this clarity of focus, which creates 3 simple but powerful growth drivers in our business, which are shown here. Capturing record reversion and adding value through asset management, developing out our attractive logistics pipeline and most recently, securing opportunities in data center development, which will provide the potential for exceptional returns. Of course, the focus of today is primarily the opportunities that we have in our logistics development and data center pipelines. But before we dive into that, I'll briefly touch on the opportunities that we have within our existing portfolio, which was recently complemented by the acquisition of UKCM. Starting with UKCM. Well, we're making excellent progress. We're integrating and maximizing the value of the logistics assets that we acquired. And these complement other urban logistics assets that we've developed and acquired over the last 2 years. And as the upper chart on the screen here shows, within a short period of time, we've already added over 12% to the rent roll of UKCM through over 20 asset management events, including letting up 3 vacant units, in total, adding around 6% to capital values. Now as you know, we're making excellent progress at selling the non-logistics assets in the UKCM portfolio, having disposed of more than half of these since the purchase. So we're well on track to achieve our objective of being back to 100% in logistics by May 2026. And we expect to achieve disposal values above the effective price that we paid for these assets. The industrial logistics assets from UKCM plus the other urban and last mile assets enhance the significant rental reversion opportunity that we have within the portfolio. And as a reminder, rental reversion reflects the difference between our contracted rent and the current estimated market value of our properties. And as Henry will touch upon shortly, rents in the U.K. are continuing to grow, supported by the attractive fundamentals of our market. Now in most cases, capturing that rental reversion is just a matter of time until the next rent review or lease expiry and typically requires no or very limited capital. And we have a really great track record at meeting or exceeding our estimated rental values, typically capturing anywhere between 100% and 104% of ERV. You can see on the right here that the rental reversion alone and letting up a vacancy in our portfolio gives us the opportunity to add nearly GBP 80 million to contracted rent. And we anticipate capturing nearly 80% of this in the next 3 years. So we show here in the gold bar on the chart, you can see combining with the other 2 drivers gives us the opportunity to significantly grow rental income over the medium to long term. As we announced last week for the first time, it's these opportunities that provide the potential for us to grow adjusted earnings by 50% by the end of 2030. Now the chart gives an indication of the contributions that we can expect from asset management, from logistics development and from data center opportunities. And I'd also like to note, as you can see on the right, that these figures do not include any future market rental growth, asset management initiatives or potential investment acquisitions or indeed the 1 gigawatt of further grid connections that we have within our data center pipeline. So to conclude my opening remarks, we've outlined here how our investment case offers investors attractive returns, supported by our 3 growth drivers. And this growth is underpinned by quality and efficiency. Quality in our modern sustainable assets, supportive long-term markets and our world-leading clients and efficiency in our lean and agile structure, triple net leases and our strong balance sheet. Now these combine to drive earnings growth, dividend progression and the potential for further capital appreciation. And I strongly believe that we have all of the necessary attributes to be successful in the short, medium and long term. And I hope that today's session leaves you with the same confidence. I'll now hand over to Henry to run you through the supportive fundamentals of the development market. Henry?

Thank you, Colin. Good afternoon. I'm Henry Stratton, Head of Research at Tritax. Over the next 15 minutes, I'm going to talk about the market environment with reference to our development activity. I'll cover 3 key structural trends underpinning demand, why U.K. demand remains resilient despite ongoing global uncertainty and has the potential to pick up further, how supply has reset and what the future pipeline looks like and why we're confident that logistics rents will continue to grow. As I do this, I'll highlight some of the market detail that our development business is positioned to capitalize on before handing over to Andrew to talk about development in more detail. We'll then be happy to take questions. So 3 global structural trends, combined with challenges in bringing forward new supply, support the potential for enduring rental growth. The first demand trend is shifting consumer behavior. The U.K. is a leading global e-commerce market. Online-only operators and the shift from traditional retailers to omnichannel are driving significant network realignment. Second, supply chains are in transformation as businesses seek resilience as well as efficiency. Stock levels and supply chain visibility are in focus. Warehouses play a vital role from storing goods to tracking movement and technology adoption underpins all of this and buildings must be able to accommodate it. Third is sustainability. Energy continues to be a priority, while power resilience is also climbing the agenda as technology, automation and increasing use of electric HGVs impact energy needs. Now moving to supply on the right. Andrew will talk about all of this in much more detail, but quickly on planning. In contrast to some other parts of the world, the U.K. has a national planning framework that's implemented locally. Navigating inconsistent implementation and highly engaged local communities often takes considerable time and requires significant expertise. Indeed, our portfolio of land options held in a capital-light manner has taken years to curate, and it remains a significant point of difference for us. So let's drill down into the U.K. logistics real estate market and what makes it so attractive. Now on the left, you can see the impact of the U.K.'s large, wealthy and compact population. We have a surprisingly small amount of stock per capita given our market maturity. So big picture, we need the stock we have, and there's no large overhang of buildings. You can see from the charts in the middle, 47% of space is less than 15 years old, and the market typically favors build-to-suit development over speculative construction. Even in recent years, where we saw more speculative development through the pandemic, deliveries are still split broadly 50-50. And as the map right shows, U.K. logistics is not just about London and the Southeast. The largest market is the East Midlands. If you want to be a leading U.K. investor and developer, you need a national portfolio of standing assets and potential development sites, which we have. Let's turn now to where we are in the logistics real estate cycle and why U.K. demand has remained resilient despite ongoing global uncertainty and has the potential to pick up further. Now there's a lot on this slide, so let me walk you through it. Our market has proven resilient through some turbulent times. Business confidence, shown in the left chart, has been volatile, yet space under offer, the gold bars, has remained consistent over the past 3 years, and we see scope for demand to increase. Economic growth, while somewhat uninspiring, is positive. And while CPI remains slightly above target, contributing to this is very strong wage growth. Despite a very real cost of living crisis, many have seen real incomes grow recently. And this, as the chart right shows, has supported a recovery in retail sales, much of which passes through our buildings. This is clearly good news for our market. While the growth rate may ease, further interest rate cuts and any reduction in the elevated U.K. household saving ratio would be supportive. And a quick word on the autumn budget, which saw labor costs rise from April 2025. While this is arguably of greater consequence to U.K. retailers than U.S. tariffs, we think it will have a positive impact on our market in time. Companies have made clear that they're turning to more technology and automation. Within the supply chain, this is best facilitated by high-quality modern logistics buildings, and we expect to see further demand for new buildings. Now the economy underpins demand, but corporates ultimately drive leasing decisions, and the corporates in our core markets remain active. The chart on the left shows average annual revenue growth across 2024 and what is forecast for 2025. As you can see, many bellwether logistics occupiers across a diverse range of subsectors are delivering impressive levels of revenue growth. Indeed, a number are active in the market at present. And buildings must be able to support occupiers' ambitions, including technology adoption, and you can see what this means in practice on the right. Buildings are getting larger, delivering economies of scale, higher to facilitate mezzanine floors and high-bay automated racking. Floors must be strong to support these build-outs and absolutely level for automation and technology. Other features such as solar PV, natural daylighting and landscaping are important as our office build-outs and amenities as tech adoption changes the talent mix. And finally, as I touched on earlier, power provision and resilience are critical. So building specifications continue to evolve, creating development and asset management opportunities that we're really well positioned to capitalize on. And Andrew will talk more about this in a moment. So turning to the actual demand data arising from this corporate backdrop. We've seen healthy levels over the past 18 months, typically in the range of 5 million to 6 million square foot a quarter as companies continue to evolve their operations and networks. And this demand is very diverse, which is a key attribute of the U.K. market. As the chart center shows, it's underpinned by high levels of retail sales, e-commerce, a surprisingly large high-tech manufacturing sector, food retailers, food manufacturers and the 3PLs that support all this activity. The chart right shows that around 2/3 of demand is typically absorbed by new buildings. That's the dark and light blue bands. This shows just how important modern high-quality space is and underscores the importance of a high-quality development platform like ours. It's also worth remembering secondhand demand and renewals on existing buildings. Although renewals are not shown in take-up, both secondhand and renewals underpin asset management opportunities for standing stock, which is a significant opportunity for us and our sector. And one last point here. For a number of years, speculatively developed buildings have accounted for a higher share of demand than has historically been the case, around 35% to 40%. That's the dark blue at the bottom. I'll touch on why this is important when we look at supply. So demand remains healthy, and we've also seen a supply reset. You see on the left in the dark and light blue bars, speculative space under construction has declined markedly and rapidly from its 2022 peak. Coming back to that 40% from my previous slide, if annual gross demand is around 25 million square foot, the market could broadly accommodate around 10 million square foot of speculative supply. And this is where the market is today. As you can see, chart left, speculative space under construction currently totals 11.5 million square feet, but around 1.6 million of this was under offer, leaving around 10 million square feet scheduled to complete in the next 12 months. Supply and demand for new space is, therefore, broadly in balance. You can also see the broad regional spread of this pipeline in the map. And as the graphic right shows, we have good visibility into the 2026 speculative pipeline. Year-to-date announcements are down 18%, and second quarter numbers were skewed by 2 very large projects, which accounted for more than half of announced square footage. So overall, the speculative pipeline does seem to be slowing further. And a quick comment on build-to-suit or pre-let activity. Projects are shown in the gold in the left-hand chart. Under construction volumes recovered somewhat mid-2024, but recent announcements have been relatively light, but we think these dynamics are starting to shift. Anecdotally, we're seeing more interest in build-to-suit, and we expect this market to pick up over the next 12 to 18 months. So let's put this all together and explain why we're confident on logistics rents. The first point is that while total vacancy here in gold has ticked up in Q1 2025, Vacancy for new buildings, the orange component, has remained steady over the last 12 to 15 months. So as ever with vacancy, composition is important. The uptick in total vacancy across the first quarter was almost entirely down to secondhand space coming back to the market. While for new buildings, the market is in effect, absorbing as much as it's being built. And of course, local market dynamics are important. While there are pockets of supply, it's not just about geography. It's also about building size, specifications, access to major roads, traffic pinch points and other factors. Building and owning the right product in the right place requires the right kind of expertise. And that's the kind of expertise, knowledge and relationships, which our development and asset management teams have. So when it comes to prime headline rents as reported by CBRE and shown regionally in the table, logistics rents have ticked higher, shown here in shaded green boxes since 2023. And as the triangles show, Savills data on signed leases suggests that the differential between prime and secondary rents is growing. Indeed, it's as wide today as at any point in recent history. MSCI data, a better metric for wider portfolio performance substantiates this rental growth with 5.2% ERV growth last year and a further 1.2% in the first quarter of 2025. We continue to believe that market dynamics remain supportive of further rental growth. And our view is that over the medium to long term, this will be in the region of 3% to 5%. Before I conclude, just to touch on capital markets, where Prime Logistics yields remain at 5% to 5.25%. Deal volumes have been relatively light this year with around GBP 1 billion transacted in each of the first 2 quarters. But capital markets are open. We've seen a number of portfolios transact and investor interest remains high. Both long-standing investors and new entrants are active in the market, looking to capitalize on those attractive entry yields and strong market fundamentals, which continue to provide the scope to grow income and undertake asset management or development initiatives. So to wrap up, the structural trends underpinning our market continue to support demand for U.K. logistics real estate. Barriers to new supply, particularly around planning, remain significant. The U.K. economic backdrop, while somewhat underwhelming, is improving and recent events have affirmed the critical role our buildings play in supporting supply chain evolution. Demand for modern space is diverse and meeting it requires a national portfolio of high-quality buildings with a high-quality development platform. Supply has reset and while market dynamics nationally look well balanced, there are pockets of both over and undersupply. And lastly, we continue to see healthy levels of occupier interest and rental growth. So if I could leave you with one thought before handing over to Andrew, is that market fundamentals are well placed to support enduring rental growth, which we believe will be in the region of 3% to 5%. Development and asset management opportunities certainly exist and we have the kind of specialist on-the-ground knowledge required to identify and execute on them. I'll now hand over to Andrew to talk more about that. Thank you.

Always good not to trip over the wires that are in front of the screen. So good afternoon, everybody, and thank you to Colin and Henry for setting the scene both for our business and for the market generally. I'm Andrew Dickman, I'm the Managing Director of Tritax Big Box developments, and I was one of the shareholders who sold DB Symmetry to Tritax Big Box in 2019. Today, it's my pleasure actually to explain the impact of the business that I run on the fund -- the Tritax Big Box REIT generally and also to focus on what Tritax Big Box REIT can expect from the development business over the coming years. In summary, the development business has the ability to double the size of the rental income stream of Tritax Big Box REIT. So in explaining what we do, I'd like to touch upon our experienced team. I'd like to talk about our land portfolio. I'd like to talk about our land options, our relationships, our planning capability and the way that we deploy capital. We have a very experienced team of 45 professionals who operate out of 3 locations. That locational reach is fundamentally important. It makes us different to a lot of our competition, being able to do business and have relationships that are deep and meaningful in different geographies gives us a USP. We also have a track record of talking to our customers, the occupiers of our buildings, to understand what they need, what their insights are, how their businesses are evolving so that we can supply to them a product which is long term in its use for them and cutting edge in its delivery. Our land portfolio, as Henry noted, and Colin mentioned as well, is the largest logistics-focused land portfolio in the U.K. It's taken us decades to build. You can't just rock up a farmer's farmhouse, pop in through the front door and come out with an agreement to buy land. Our reputational knowledge and our reputational protection is hugely important in this sector. We're also focused very, very carefully on key locations. Almost 80% of the land that we control is controlled by way of land options. The land options that we've developed over decades are quite complex, but they're also very fair to us and to the landowners that we contract with. The land options give us long-term control in a very capital-efficient way and enable us to control and manage our development profit. They're embedded with discounts and they're drawn down in tranches generally. I've mentioned the word relationships a couple of times already in this presentation. We fundamentally believe that the relationships that we have with the landowning community, with our customer base and with local and national politicians is meaningfully important. We take care of those relationships. We polish them, and we make sure that we do the right things always to make sure that those relationships grow. It's a thoughtful and experienced approach. It's also very long term. Our planning capability is second to none. We're delivering very large schemes and the planning consents for those very large schemes. We have to be thoughtful about how we do that. We have to make sure that all the stakeholders are considered and not taking something away from the project. We secure our consents through very detailed and comprehensive stakeholder engagement. Once we've got planning, it's time to deploy some capital. We're very asset specific. We do development on a pre-let basis and on a speculative basis. We measure each process carefully. We take our underwrites to main board with a thoughtful approach, and we generally beat the outcomes of those underwrites. We're agile in our decision-making, and we are very, very careful about how we deploy capital. By using these skills and these tools, we're able to provide superior risk-adjusted returns. So in summary, we reduce our risk through land options, through our planning skill sets and through a considered and precise capital deployment. We maximize our returns from the entry price through our pace of delivery and through growth in rental income. And that's how we are able to supply product to Tritax Big Box REIT with a yield on cost of between 6% and 8%. There's a lot going on, on this slide, so I apologize for that. Let me just drag you through it from the left to the right. The strong track record that we've created over the years that we've been part of the Tritax business has allowed us to provide GBP 71 million of annualized rent. That's come in 9.1 million square feet of lettings and an additional 2 million square feet of freehold sales that we refer to as DMA, Development Management Agreement contracts. We've been very agile in our delivery of spec versus pre-let build-to-suit. You can see, for example, in 2020, very significant spec program, that was the drive to supply product into the COVID boom logistics market held. We then geared that back until 2024 when actually 100%. So 90% of our spec lets within 12 months of completion of the building and much of it lets whilst we're constructing. 64% of what we do is pre-let, however. One of the nice things about spec is that from the underwrite that we go to board with, we've been able to, on average, improve the rental tone by 15%, which improves our returns. You can see in the bottom middle slide where we've moved our yield on cost from 5.8% in 2020 to 2024 when it's 7.1%, 6.4% is the average annual yield on cost return. I'm really quite excited about the slide at the bottom of the right hand of this slide though, because whilst we've been doing all of these things, delivering all of these buildings, all of this income, we've also managed to take on enough land to completely replace the land that we've eaten through. That gives our business a really genuinely evergreen approach to its land assembly, and it gives us the future to mean that we can double the rental income of the fund. So I mentioned our team at the start of this. We're nothing without our people. These projects don't get off the ground without the people to deploy them. We've got a very stable settled team of seasoned professionals. Obviously, we've got a lot of developers. We're a development business. But we internalize town planning and construction risk management and finance, and we have a strong analyst capacity. The average of 23 years for our senior leadership of experience is really quite important. It's generational. And these individuals have been with the business for a considerable period of time. And as I say, split across 3 different geographies, which I think is meaningfully worth knowing. That group of people has been able to curate 25 sites across the U.K. We've been a little bit more than that and a little bit less over time as we've chewed through the developments. Those sites can accommodate 38.6 million square feet of property development and have the scope to add, on current numbers, GBP 320 million of annualized rent. And they're spread around the U.K. But it's really important to note that we discount 90% to 95% of the sites that come across our desks. We're not in the game of taking on properties unless we absolutely believe in them. We start by managing the planning and political environment in a location. And being simplistic about it, if we don't believe a local authority wants us to develop in the authority, we won't come to town. We're very anxious that our sites are strategically connected in every way, shape and form, not just by being on motorway junctions. We're very anxious to make sure that there's labor availability and the right labor. As Henry mentioned, the talent pool and the talent mix is changing in these buildings and access to it is really important. We also work at size and scale. Our schemes are generally very large. It's unusual for us to come to board with something less than 1 million square feet. That will mean that the scheme will have multiple phases, and it will mean that we're in the location for a long time. We're thoughtful about power availability. Power requirements are only going up. These buildings and the sites need to be able to accommodate that. We're very thoughtful about ESG. We don't just do biodiversity net gain because we have to, we do it because we believe in it. We're not a seller of these assets. So the parks have to last over time and have to adapt to that people want to work in them. And we're also concerned with things like ground conditions, topography, how this works. If you pass all of those hurdles, then you're a site that we want to be involved with. I mentioned land options at the start of this. We take 83% of our portfolio via options. The other 17% is owned. So we do buy land where we think that we should and where there's an opportunity. But of the options, the 83%, we believe they're really, really good for us and importantly, for the landowners who own the land. So these options are long, they're 10 to 15 years. That lets us have 2 or 3 cycles of the town planning circuit, and it means that we can time our run in the most effective way. This land is priced. The logistics market value at the time of drawdown is what we pay less a discount. That protects our development margin and gives us some recompense for the skill sets and the knowledge that we deploy onto the sites. We deduct professional fees and infrastructure costs from the purchase price. So we're basically buying serviceable logistics land that can then be developed. And we pay an option fee. So the farmer gets some money day 1, but not enough money to make a dent on our balance sheet. So it's really, really cost effective. So the benefits for the landowners, they're working with a trusted expert. They can have huge confidence from that. They can continue to farm the land, so they're using it for free. They receive an option fee, which is meaningful to them. They don't have to put any further capital or time investment into the process. We'll do that for them. And when we're successful, they get a significant payday from the uplift in value that we create. It's good for us because we've got security and control of the land. We're capital efficient. So for our balance sheet, it works very nicely. We've secured a discount, so we don't pay full market value when we draw the land down. It's highly flexible. We call the tune with time, and it removes long-term land pricing risk. It's long-term and positive relationships, which are critical to securing and successfully delivering these options. So what's this slide telling us? It's telling us what happens once we've deployed an option. So we've drawn some land down. What do we do next? The next thing we do is make this land credible. It's no good having a field and saying to an occupier, we can build you a building, and it will be there in the next number of months. They've got to be able to see that, that is real. So we infrastructure the sites. So when an occupier turns up on site, it's very clear that the site can be delivered. What we can do then is find our way out of the schemes in 4 different ways. Now I'm going to just jump to the bottom of the screen where we say sell land because it's the one that we don't do very much. So we can -- after we've drawn the option down, we will have a planning consent. We will know what it costs to service the site, and we could sell it if somebody wanted to pay us enough money to compensate us for the loss in development profit. It's not our preferred way of doing things, but it's a credible solution. More commonly, the 3 build-to-suit speculative or freehold turnkey DMA routes are the routes that we deploy. Build-to-suit is our preferred, 60% plus of our delivery is build-to-suit. Why do we like it? Generally, it's larger buildings. The letting risk is completely removed. There's certainty of returns. Its only downside actually is it can be slow to negotiate. An alternative is speculative development. We do this because it puts our scheme on the map for agents and occupiers and because in our opinion, it's the right thing to do for the project. It optimizes market requirements. We have a specification, which we believe is market-leading and on a fairly recent tour around our competitors' sites, I can confirm that it is market-leading. It's very clear to an occupier when a speculative building is delivered that they can operate from it almost immediately. And you would be surprised how many times we have occupiers who come to spec buildings, knowing that they don't have time for a build-to-suit solution to occur. There is risk in letting. That risk is managed through our process, and I refer you back to the fact that for our spec portfolio, 90% let within 12 months of PC and many of the buildings let in build. And the final way through this, and more to follow on this is the freehold turnkey development management agreement. It limits the capital that we need because we use the freehold purchases money to build the building. It captures freehold demand, noting that manufacturers are generally requiring freeholds and last year was the largest uptick in manufacturing requirements and sales that has been on record and that the government wants to promote defense manufacturing, we expect to see more DMA type arrangements. It supports our broader scheme development, and I'll come to that on a later slide as well. It elevates returns profiles. Its only downside is it forgoes rental income. Once we've agreed to do one of those routes, not the bottom one, obviously, we complete the building, generally 12 to 18 months of construction against a fixed price contract. So the construction risk is very carefully managed. These short build times reduce market risk. We are net zero in construction. We were the first large-scale logistics developer to publicly state that they would do that. And we aim for BREEAM very good and excellent where we can. So the quality of these buildings is right at the Zenith. I've trailed development management agreements a couple of times in this presentation already. A live example is a good example. We were approached by Siemens Healthineers, huge business based in Oxford with 2,000 employees in the town, making medical instruments. They wanted to buy a building, and they wanted to buy a building where they could relocate their staff or they were going to buy a building in China. My colleague, Tom Leeming, settled back over there, managed to convince them that it will be a good idea for them to take a site from us. And we agreed a deal whereby they bought some land from us, and we built a building for them that we handed over last month. It's 371,000 square feet. So it's quite a significant manufacturing facility and all of their staff are being retained in the U.K. and moving over to it. That's not the whole story. One of the motivating factors for us was that when we went in for planning for Siemens, we went in for planning for another 1.8 million square feet of development and got consent for it. We used almost a Trojan horse type approach to get through the planning gate. It's a really attractive thing for us to be able to do. We get early profit because we sell land. We then use the occupiers' money to build the building, and we take our profit at practical completion. So it's cash neutral for us in construction. These sort of flexible approaches are really, really helpful to us in delivering our schemes. So a little bit of a progress update on some of the things that we've got on site now. In 2020, we speculatively developed a building at Kettering, which is now let to Iron Mountain, 300,000 square feet. In doing that letting, we infrastructured the whole rest of the site. And over the course of the last 12 months, we've been able to agree a letting to a pre-let to a large Internet retailer who I can't name, choose an NDA, but you can probably guess. That building is under construction. Steel are up now. At the same time, we agreed a DMA approach with Greggs, the high street retailer, and we're building 350,000 square feet for them. Those 2 buildings will finish during the course of 2025. At Biggleswade, we agreed a pre-let with Warburtons, the bread company, famous for Christmas adverts as well. And that is part of a 930,000 square foot planning consent, which has been granted, and we're on site building all of those buildings now, talking to occupiers. And lastly, for Progress, 273,000 square feet built in Merseyside, PC last month. There's actually a viewing there this afternoon, so I'm quite excited to find out what happens with that, with our buildings being marketed and that is spec development. In addition to that, we've got 0.9 million square feet of pre-lets in solicitors' hands as we currently speak. We have 1.5 million square feet of speculative units in negotiation, and we have 1.9 million square feet of pre-lets in discussion. So I've got a little case study for you, which hopefully you'll enjoy, which is our Biggleswade scheme. And what this will explore for you is how we do multiphase development over long periods of time successfully in a flexible customer-faced manner. [Presentation]

Quite powerful, I think, the short video of the project. Again, just to sum up, we're delivering superior returns, and we have the potential to add over GBP 320 million of income to the Tritax Big Box REIT over time and at a development yield of between 6% and 8%. How are we doing this? We're doing this because we have a very engaged, motivated, experienced team who are able to deploy their knowledge and understanding of long-lasting relationships. We have a large and capital-efficient land platform, and we're very careful and agile about making our decisions and precise in how we deploy capital. That's how we're delivering between 6% and 8% yield on cost. In the near term, that's buildings starting to be constructed within the next 36 months, we have the capacity to produce GBP 1.28 million extra rental income. And longer term, beyond that time frame, another GBP 195 million based on today's numbers. So thank you all for your attention. It's appreciated. And I'd like to invite Colin, Henry, Frankie and Ian to sit on the stage here, and we'll take some questions if you've got any.

Thanks, Andrew. Just to mention, so this Q&A session is primarily focused on the presentations that you've heard just now from Henry on the market and Andrew, on our development activities. There will be the opportunity to ask questions relating to our financial position, balance sheet, et cetera, from Frankie a little bit later and also in relation to the data center activity from Tim O'Reilly in the second Q&A.

Just to remind [indiscernible] on the webcast, you can ask your questions on the chat box for as long as my laptop has got battery, which is probably about 35 minutes [indiscernible]. Please do, far away.

Can you hear me?

Yes.

It's Andrew Sowders from Shore Capital. Clearly, you're on a journey. The company since it floated in 2013 is no longer just big boxes. I wondered if you could perhaps just outline to us what the sort of optimal portfolio looks like going forward in terms of asset allocation. Clearly, your exposure to last mile with last year's UKCM deal is a growing importance, the data centers development we heard about last year. So perhaps you could just talk us where the portfolio moves forward in terms of allocation and perhaps how capital recycling plays a part in that. Perhaps to just touch on how big box is less important given your interest now in data centers and last mile. And then I'm going to ask another question, which I appreciate you may not be able to answer. But if you could tell us the parts of the warehouse portfolio that you were attracted to and perhaps what elements of that portfolio might be considered noncore and up for disposal?

Thank you, Andrew. Very useful question. So look, I think the first thing is we're not changing our DNA. It's in our name. We truly believe in large-scale logistics assets in the U.K. We think that market has got a long way to run. And as we mentioned earlier, that provides the foundation for all the other wonderful things we're doing in our business with really high-quality buildings, the great customers, long leases. And you've seen -- if you look back, we had a 10-year track record of 100% rent collection. I think that speaks for itself. It's probably a world record for a company of our size. But in addition to that, we are staying laser-focused in terms of our business strategy in capturing opportunity within areas that touch and affect our business. Now there was a time when yields were too tight for us to venture into the urban market. We were, and as Andrew mentioned, we've developed smaller scale buildings as well to complement our portfolio to provide our customers with a broader range of service offer. And UKCM gave us a perfect entry point. And I think we bought very, very well there, and we've talked about all the wonderful things we've been doing and the performance it's already provided to us. The Warehouse REIT opportunity is another similar approach to UKCM. If we are successful in that, it would take us up to around about 20%. We're not setting ourselves any hard targets on that. It will be opportunistic, and we look at the balance of that on a risk-adjusted returns basis. But we don't see these elements. I mean, data centers are another really, really exciting component part of our business when you think about the yield on cost and the returns we can deliver from that. But we don't see these elements as being mutually exclusive, okay? We believe that we can do every single one of them without that being to the detriment of the other. And this comes to your point, Andrew. And part of the reason why we are so confident is we have a really strong balance sheet, and we've proven time and time again our ability to transact by selling really high-quality liquid assets in the market. We've done that time and again in some of the most challenging markets. Now that market is actually improving right now. So that's what gives us the confidence. And hopefully, that answers the first part of your question.

Matt Norris from Gravis. Great presentation. On the developments on the yield on cost on Slide 25, you showed the yield on cost progressively getting better, so from 5.8% in 2020, then 2024, 7.1%, you're giving an average of 6.4%. And then the guidance is 6% to 8% going forward. What enables you to improve this yield on cost?

So what we've seen across the period, and Henry noted this in one of his slides, there's been quite a diverse set of volatile occasions that have occurred. What we've been able to do through the piece is iron out the volatile bits. So if you look at today, construction costs are very stable, but rents have continued to grow. So our yield on cost profile has improved through really, really active risk management against a growing rental market. So they're the 2 key ingredients there for us. And we expect rents to continue to grow between 3% and 5%. And we also expect, and I should touch wood when I say this, shouldn't I, that if there is another volatile event that we have the skill sets to manage it and keep those risks within type profile.

It's Tom Musson at Berenberg here. Appreciate the 3% to 5% ERV growth guidance. Just be interested to know your view on sort of relative rent growth prospects on the larger boxes within that, just given the changes to business rates, in particular, the large building multiplier change? That's the first question. Then Andrew, perhaps a second question for you. You showed a chart showing the land bank has sort of more than been replenished over the last 6 years or so. Just wonder how does the quality of your land bank today compare in your mind to what it was 6 years ago, perhaps in terms of the micro location and rent potential?

So can I take the second one first? We haven't changed our assessment criteria with the land. So noting that we've got a very significant land bank we don't have a target of keeping it a very large land bank. We have a target of keeping it the very best land bank. So where we've taken on new sites, it's the response to gaps in our network or a really good opportunity coming along that meets all of our criteria. So from a quality perspective, it is at least as good as it was 6 years ago. And you could argue that some of the criteria have actually improved. So I'm very confident about the quality of our land bank. In terms of rental growth prospects, Henry noted the supply/demand imbalance that we have in the U.K. He also touched upon the inherent problems and challenges of getting town planning consents in the U.K. Nothing is going to change in those 2 things. So there remains a constraint and a pressure upon actual delivery of land and buildings into the market. And we think this underpins the growth criteria of rents within the sector. And I don't think that changes outside of bigger boxes. Those bigger boxes are, in many respects, the core assets within a lot of these businesses, portfolios of logistics assets. And as such, I think that underpins the likelihood that rents will continue to grow in the larger assets in the sector.

Just to add to that, as you talked about the critical element of this. And I think this gets slightly forgotten just how important these large buildings are as that first element of the supply chain. When occupiers are looking at it, they're looking at the full cost of operations out of it and some of the improvements and gains that they're able to make within their operations, particularly around technology, et cetera, which we know puts into place new requirements in terms of the quality of these developments. So we're seeing that shift towards large buildings within take-up. It's still, as you say, very difficult to bring large projects forward. So you don't have the same level of supply certainly on the speculative side. And these are really critical components of what companies are doing. They're putting CapEx upfront to massively improve their operational performance within them, which is then differentiating their business. So rent is only a small proportion of that within it. It's important as are rates, of course, as well as employment. But all of these things are coming together to mean that these are really critical pieces of that jigsaw to serve this kind of ever-evolving U.K. network that we have, the need, obviously, to go far further and wider in terms of delivery. It all stems from this first mile infrastructure, which is critical.

There's a piece of research being done and refreshed recently about the amount of warehousing space that's generated by building a new house. So noting that the government has set a target of 1.5 million new homes in the U.K. during its term, 69 square feet of space is required to support each one of those houses. You can do the math just as well as I can, but it implicitly supports growth in the sector.

Tom Walker from Schroders. Can you just talk to me or tell us a bit more about the yield on cost, the industrial versus the data centers. Both the figures like-for-like in that are you using purchase price of land? Are you using written down value of land? Just want to understand the difference between the 2, please.

[indiscernible] for me. So just to recap on the guidance, I'll come on to touch on the DCs, but logistics is 6% to 8%. Data centers is now 9% to 11%. The methodology or the basis is exactly the same. It's the yield. It's the rental income that will be derived come lease take-up over the cost of delivering the site. So cost of land, cost of infrastructure, cost of construction activities.

So it's historic cost of land is what you pay for it.

Correct.

And just a question from the webcast. Does that yield on cost calculation include all the legal and professional fees as well?

It's absolutely everything, including incentives if they're required.

And sorry, one more, if it's all right. Just on the 1 gigawatt. Can you just talk about whether that is kind of ready -- that's contracted, ready for kind of use tomorrow if you found a hyperscaler, kind of how the drawdown works on that, how real that 1 gigawatt as opposed to you get a few megawatts in 3 or 4 years' time?

Do you mind if we answer that [indiscernible] steal Tim's thunder, who's going to talk exactly to that point.

Callum Marley from Kolytics. Andrew, I think you mentioned how the new defense spending policy is increasing manufacturing demand and potentially turnkey solutions. How big do you see that opportunity? And then what level of returns do you have to pencil to get involved there instead of build-to-suits or speculative developments?

So it's an interesting question, and it's interesting because we don't have a definitive answer to it. The defense spending review clearly is going to throw more demand into the sector. We are seeing inquiries already from defense manufacturing organizations. We also think that there's going to be a degree of bias towards the government spending money with U.K.-based organizations providing U.K.-based defense products. I can't foretell the future in terms of how much space that's going to require. But I think it will become clear over the next 12 to 24 months as the cogs start to move in the MOD and we start to see the evolution of the government's policy change.

I'd just add to that as well around the wider manufacturing shift, 26% of demand last year to manufacturing, I mean, record high in terms of how active that sector is really looking at making the changes that we've seen across other sectors as well. Defense is now an additional component of that certainly in comparison to even a couple of years ago. So that element and the way that, that transact in the market, which Andrew has talked about is really important because it's clearly going to be a driver for some time around those sort of multiple aspects of defense reshoring and how that plays out specifically in the U.K., which is a little bit different perhaps to other places, but this is all creating additional new demand from different and other occupiers.

[indiscernible] In the context of the quality bifurcation with new build vacancy rate below its peak in early 2024, what is the risk of secondhand stock becoming competitive against post refurbishment? And how do you account for that risk?

Yes. I'll take that. I think it's really important. There's room for all of this within the market. So back to those high-level stats to kick it off just in terms of where the U.K. market is, there isn't the big overhang. So different buildings playing different roles for different occupiers is the important thing here. And often, that means an element of kind of covenant evolution, I guess, into that secondhand space. But also a lot of businesses don't necessarily need those same drivers that we've talked about. There's multiple reasons and ways that you use different space. So I think really important with this, we've focused on the development side of it in the presentation, but we're seeing very high levels of renewals in the market at the moment. That secondhand space typically does about 30% of demand as well as companies move. And so there's plenty of scope within there for companies, both nationally, regionally and locally to drive the requirements around that space. So I think both elements have a really important role to play, but we want to be able to operate between the 2 of them is the key.

That's actually benefiting us significantly in terms of asset management opportunities with occupier renewal in our buildings. And obviously, when the buildings are large and important, then that's a key win for us in terms of retaining that covenant for a new longer-term lease.

And just probably a question for Henry. How much of the Q1 '25 increased demand for omnichannel and manufacturing reflects seasonal activities versus a structural change in demand?

Yes. So first quarter, without giving too much detail, was a little bit nuanced in terms of the composition. There was a couple of very large deals, which supported the overall [ 5 million ]. I wouldn't really point towards a single quarter in terms of driving these numbers from different areas. I'd look at the slightly longer-term elements, certainly 12 months, 26% manufacturing has talked about, 3PL is a little bit less active over the last 12 months, but they're very much in the market right now. A lot of retailers turning to them in the short term to solve some of their current needs as they kind of reflect on the changes that came from the budget, what it means to their business and how they might evolve that going forward, which, as we talked about, means more technology automation is likely to drive the requirements for new buildings. So I think within that, we certainly see online-only operators back in the market. Retailers, I think just reflecting a little bit on the news from last autumn and the implications to them, that [ 5.6 billion ], which I showed on the slide, and we expect that to continue to drive this diverse market environment, which is so important to the U.K. logistics market.

We had a couple of questions on funding, but I think we'll hold those back until you've had a chance to speak, Frankie. Are there any further questions from the room before we move on to our next section. Great.

Thanks. So I'd like to invite Kevin to come and talk about the data center market.

So good afternoon, everyone. I plan to introduce myself, but I think that's probably the third time I've been introduced. Nevertheless, I'll describe a little bit more about my role so far or at CBRE, I should say, and I lead our data center research and consulting practice. So I follow track report on London data center market, the London data center market and all data center markets across Europe for a living. And so I'm here today to talk to you for the next 15 or 20 minutes about the London data center markets role in the world, let's say, how big it is, all the kind of key questions you have, you can imagine. And so with that, as the title suggests, it is obviously a prosperous market in place. And I would love to tell you just how big it is and how fast it's growing and why it's so important. And so I'll do so. But before I do that, I think it's important to talk about what a data center is before. So I've been with CBRE for about 3.5 years now and have been employed by a data center provider before. And so it's easy to -- with the curse of knowledge to assume that people know what you mean when you say data center, especially those that are real estate investors like yourselves. And I'm continually surprised by how few people know what a data center is or only have a vague notion of what that is. So when I'm talking about the London data center market, first and foremost, let me describe what I mean. And it is turnkey facilities, first of all, so I'm not talking about powered shells or not even talking about self-builds, although self-builds matter as well, too. But the part of the market that has really grown and propelled London to the heights that it's at now is I'll point to the screen here, the wholesale colocation bit. So London has always been the largest colocation data center market in Europe. And up until recently, the second largest market in the world, it's now third behind Atlanta, Georgia in the United States. But let me put that aside for a moment. But the market has shifted from one to that is a retail focus. So what do I mean by retail? I mean you can read the screen as well as I can. And so that's smaller data centers essentially that are typically built for enterprises, all right? And the deal sizes are smaller. But the market over the last, let's say, 8 to 10 years has absolutely exploded in London and no hyperbole there at all. And by that -- and that's because of a term that you may or may not refer to, and I will say ad nauseam throughout the presentation is hyperscalers. And so that's just really large American tech companies whose names you're well familiar with, household brands that I'll talk about in detail that need vast amounts of compute power, and that means data center space in real estate terms or data center capacity, as you'll hear people in the industry refer to more. And so that market has transformed. It's morphed itself into the London data center market rather, it's much more wholesale focused, although it's growing in conjunction with hyperscaler self-builds. So you'll hear -- you may have heard of Google's development in Havering. And there are other self-builds that dot the landscape. And in investment terms, funding capital markets, there's often a question about how the growth can be two-pronged. And but the fact of the matter is, it is. So the wholesale colocation market is growing and outpacing the self-build market. And it's doing so because of the vast need for, again, compute power. So hyperscalers, so Microsoft, I'll steal some of my thunder here; AWS, Amazon Web Services; Google, primarily in Europe, can't build data centers fast enough. They can't get secure the data center space they need in a timely manner. So they turn to the wholesale colocation providers for space. And so those 2 parts of the market are growing fastest. And they are -- but one is not growing at the expense of the other. So I want you to take away that, again, the point that -- well, before I go to that, London is the largest colocation data center market in the world. There are significant constraints though, in the market. So power chief among them and available land. I'll talk about that a little bit further. And then finally, but there's still incredible demand. And so I've already touched on the demand point. So you'll hear me revisit those 3 points throughout. I thought I had a content pages in here, which is why I sort of skip ahead. So anyway, with that in mind, so going back to the point about the size of London. So you'll hear me or people in the data center industry mention London and the U.K. interchangeably. And that's because, again, London is the largest colocation data center market, data center market, period, in the world. And so what that means is there is -- the 2 are synonymous. So London has in 2023 terms or '24 terms rather, there about 80% of the supply that's operational in the U.K. can be found within the rough boundaries that we call London, right? And so Manchester is a blip. It's a small second market. Many in other European countries, there are secondary markets that matter. Germany is a great example. So Germany is the second largest, and you'll see Frankfurt and Berlin and a number of markets that are more akin to one another, let's say, although Frankfurt is easily the largest. U.K. is not like that. London is by far the largest. And so the development has been very much in the Southeast over the last 30 years. That time frame, I mentioned for a reason because that is how the colocation data center market has grown or that's the time frame they're in. And so again, that is largely a function of retail, at first retail growth. So enterprises, big banks who used to sort of command, hold the balance of power, negotiations or take up most of the compute space, the data center capacity in London. Those days are long gone. In fact, the enterprises are struggling to secure the capacity they need from the data center providers of the world. So what that means is that there's a key growth driver there, and I've already talked about it to some degree, and that is the cloud computing growth driver. So if anyone in the room uses Microsoft OneDrive or Google's cloud service, those are examples of consumer cloud services that are very much leading to the need for hyperscalers to find the vast amounts of compute power that capacity that I've mentioned before. All these other drivers are very much in place and are very much required or leading the market to new heights. But the fact of the matter is cloud computing and less so AI. AI is now artificial intelligence that has emerged as a driver. And so these data centers that house, process, transmit data, all the telecommunications kit, all the servers of the hyperscalers are more required than ever. And the problem is that there simply isn't enough. Well, before I get to that. So increasingly, the advancements in technology and some regulation are drivers as well, too, but cloud computing is far and away the #1 growth driver of the London data center market and the data center market in Europe and the world around for that matter. And so that's led London because of the fact it's the financial hub of Europe, of course, and now the hyperscalers descended upon Europe, if you will, set up what they call cloud regions, which are comprised of availability zones. Those availability zones need equipment to power them. So availability zones are comprised of servers and like I say, all sorts of other tech kit that goes into it. And so that hyperscaler investment in London has led London to become the #1 market, as I mentioned already in Europe. And so you -- that there really is a bifurcation or a division, a segmentation of the top 5 markets in Europe, which are often called FLAP-D, that's a term we coined at CBRE years ago. And most of the new supply, 70% of that new supply will be directed towards the FLAP-D markets. So Frankfurt, London, Amsterdam, Paris and Dublin, of course. And that will lead to about 75% of the operational supply in Europe to be in those 5 countries or 5 metro markets, excuse me, by the end of the year. So as much -- you'll hear a lot of talk about secondary markets in the data center world, so cities like Milan, which have grown very quickly. Monstrous growth rates, we're expecting nearly 70% -- or I would say -- well over 50% supply growth in Milan. So London isn't growing anywhere nearly that rate, but it's a much larger market. And so these secondary markets get a lot of attention. But the fact of the matter is, again, most of the operational supply will be found in those top 5 markets. So London and Frankfurt are kind of the super league, if you will, data center markets. London, by the end of the year, will account for about 22% of the 5.5 gigawatts that we expect to be operational. And a distinction I probably should have made before is we talk in -- not in square meters. In the data center world, it's power, again, because of the compute intensity. In other words, the power required for all the kit that I mentioned that these data centers house. So less square meterage and more power talk. But by any measure, London is, again, easily the largest data center market, as I've said already, Frankfurt closing that gap. But London almost remarkably equaling the secondary markets, closing that gap. So all the secondary markets we track in Europe are almost -- there's only a narrow gap now between London and those secondary markets. So increasingly important to be in London for all the good reasons that you see before you. Unfortunately, though, so that network effect and that data center development that has driven the London market to the heights I've mentioned is has meant a lack of power and land availability. So I think one of the earlier speakers mentioned how data centers are critical national infrastructure in the U.K., the 13th sector to be anointed as such. And that is a reflection of how much data center development has happened and how dependent we are on data centers to enable our lives. Of course, the world is only becoming more digital and data centers are the backbone of that effort or that dependence that we have. And the grid simply can't sustain itself. It's a first come first serve sort of power allocation system here in the U.K., as you may already be aware. And so if you're looking to develop data centers in West London or anywhere in the U.K., but certainly West London where Slough resides, of course, that we would count Slough as part of. That is the largest data center cluster in all of Europe for that matter, not just the U.K. There isn't any power available really for new schemes, not for many years at least. And so the substations that developers want to connect to, if they do even happen to have land in West London, won't have power at the Iver B or Iver substation, which will be upgraded 3 to 4 years' time, depending on who you talk to. And then there are subsequent upgrades that will come or at least one subsequent upgrade. And there isn't any available power. So that means data center development is going to have to diversify. And so we're already seeing -- and by that, I mean, more developments, more data centers outside the boundaries of London. And equally, finding that land though is difficult as well to find that power and land that's scalable to whet the -- not just whet the appetite, but satisfy hyperscalers is in short supply as well, too -- so planning, it's becoming more difficult to become a data center developer only because of the time involved and the difficulty securing that power and land. So those that have the power and land have the raw resource and are potentially sitting on land, literally speaking, that can be turned into lucrative developments, lucrative data centers for hyperscalers. So an increasingly rare commodities that are only risen in value. And so as a consequence of all that great data center development, all that data center construction that I talked about leading the markets to the heights that they've reached is falling availability. So that's a natural sort of -- it's a knock-on effect of all this data center construction that I mentioned. And we are now at the end of this year, will be sub-8%. And that might seem high if you're looking at vacancy rates comparatively to other asset classes in the commercial real estate sector. In fact, that's a historic low. So by the end of this year, that London rate that I'm showing you on the screen will be particularly low and unusually so. So that's -- and in Slough, I mentioned where the data center concentration is that cluster, it will be sub-5%. So the wholesale data centers that require a lot more power and larger data centers for that matter are even in scarcer supply. So that's, again, a natural knock-on of that effect. And that's again because hyperscalers reside or have developed their availability zones that I mentioned in the clusters like Slough. So less data center space to go around, and that is leading to rising rental rates. So the economics of the industry are such that there is, again, less availability. So the economics of the industry are working against the infrastructure buyer, better time than ever arguably to be a developer operator. And build costs are on the rise as well, too. So those 2 factors, the economics of the industry, the supply/demand imbalances, which we're seeing are kind of the new norm, not just in London, but across Europe, combined with the, again, rising build costs are leading to rates if you're procuring capacity on a multi-megawatt basis as a hyperscaler, representative example here of, say, 15-year terms. So looking to procure -- again, you're in the market for a lot of capacity and you're a hyperscaler, you're going to pay as much as [ 130 ] for capacity at a turnkey facility. So as I said before, I've been with CBRE for a number of years now. 3 years ago, I talked to you about rates, given the rough terms that I just mentioned in triple digits, that would be very much the exception, not the norm. So there are a few markets that have reached the heights that I've mentioned. Frankfurt is among them. Milan is a comparative example. But again, a smaller European city that you can see where developer operators are commanding less. So super prime real estate data center space in London commands a premium because of the scarcity factor and the difficulties delivering data center space that I've mentioned before. So I'm approaching the 15-minute mark, and I think I will leave it at that and hand the control over to Tim. Thank you.

Thanks, Kevin, and good afternoon, everybody. I'm Tim O'Reilly, Director of Strategic Power at Tritax Management. I joined Tritax about 3.5 years ago as part of its investment in developing leading expertise in power. I joined from the National Grid, where I was Head of Strategy and Innovation, and I was responsible for various projects, but most notably designing the future energy system to reach net zero carbon emissions. In our 4-year journey here at Tritax, we've developed what we believe is a truly best-in-class power capability for a real estate asset manager. As outlined on the left, you can see our journey began with addressing the key power challenges faced by occupiers. To put it simply, we look to match generation and transmission infrastructure with intensive users of energy, be that a large sophisticated and highly automated logistics building or as we'll explain in more detail now, the data center. I'll also explain how we have developed our data center pipeline in partnership with a leading European utility company. We're pleased to be able to reveal today that, that partner is EDF. It's the combination of the investment in our team and its power knowledge and the market strength and experience of our joint venture partner, which gives us the potential to deliver exceptional risk-adjusted returns to shareholders from our data center pipeline. When we talk about data centers, we're talking about the critical national infrastructure that underpins our day-to-day lives. But let's start with what data centers need. This is best distill down to power, proximity to other data centers and connectivity to data networks. Starting with power, large-scale modern data centers of the type that high-quality operators want to occupy need significant amounts of power. We see power requirements continuing to rise with 50- to 100-megawatt size data centers increasingly common. And to put that into context, 100 megawatts is about the same amount of power a town the size of Milton Keynes with a population of just over 0.25 million people would typically consume. Data center facilities interact with their neighbors operationally and therefore, need to cluster. And this creates the concept of availability zones. Clustering data centers together increases availability, resilience and scalability. The broader London availability zone is the most important in Europe with Slough, the most prime zone, second only in importance to West Virginia. It's worth noting that creating a new availability zone is challenging as you lack the network effect of already established area. The final point is connectivity to data infrastructure. Data center operators, particularly those that are providing cloud services or AI inference uses that are very sensitive to latency or the time it takes for the light to travel between 2 points, need good connectivity. From a service performance perspective, this means that the ideal DC locations are as close to the users as possible, which is why London and particularly Slough is so key. So while you might get access to cheaper sites and power further away from large populations, your latency will also increase. All these factors we take into account when identifying potential opportunities, and it's one of the ways we look to reduce risk. As many of you will be familiar with and as Kevin has outlined, there are significant constraints to power in the U.K., not so much on the generation capacity, but in the networks or the wires and cables that join everything up. We've outlined here the key reasons for this, which include geography. The U.K. grid is built around former industrial areas where heavy industrial users of power used to be based. But as we've outlined, this typically isn't where data center operators want to be. The energy transition is also having an effect with power increasingly being generated at the edge of the network rather than at the core from large centrally located power stations as it was originally designed. The U.K. planning system also constrains things by extending the time it takes to get new power infrastructure up and running. Ultimately, though, given the sheer scale of the power requirements of data centers, the existing distribution infrastructure is unable to keep pace with the growth in demand. As a consequence of all of this, and as you will have heard, wait times for power capacity in key data center locations is now in excess of 10 years. Grid reform is seeking to improve this situation, but the reality is in areas where capacity is needed most, the physical constraints need to be removed. This means that the existing lead times will remain. So how do we overcome these challenges and avoid this long delay? This is where our Power First approach is so critical. We've outlined here how our approach differs from the more traditional real estate mindset. On the left, the traditional real estate model would be very land focused, identify the location, acquire the site, then apply to the DNO for a new grid connection agreement and then join the queue, which in London and Slough in particular, could be 10 to 15 years. So following this approach, you could have a data center up and running in 2035 to 2040. But our approach is different. We start by looking for the key raw ingredient for data centers, and that's power. Using our knowledge and contacts within the U.K. power industry, we've identified existing grid connection agreements with defined time scales for power delivery. Unlike land, power is portable. So once we have the grid connection agreement, we can then look for an appropriate site. We have the ability to run cabling over several kilometers from the substation. Once we source the land, we have, in effect, accelerated the process, such as in the case of Manor Farm, where we have a firm commitment for power delivery from 2027. Following this approach means we have a data center up and running and critically for our shareholders, income producing far sooner. And as I say, in the case of Manor Garm in 2027, rather than 2035 or beyond. This Power First approach aims to unlock the significant upside by harnessing the marriage value of land and power. And together, this far outweighs the ingredients. Our partnership with EDF brings together 2 best-in-class experts to help unlock the opportunities in data centers. We and Tritax bring an entrepreneurial approach, expertise in land and planning and are credible with many years of delivering complex logistics buildings to demanding blue-chip companies. EDF is a European leader in low-carbon power generation with access to critical grid connection agreements. They bring their complementary capabilities, which will be particularly helpful in the delivery of power infrastructure. We've outlined here the nature of our joint venture with EDF. The first thing to note is that each of our schemes will have its own specific joint venture agreement with EDF, although the structure will broadly follow what is outlined here. Starting from the left, Tritax Big Box will own the land and real estate, including the data center itself. Big Box will receive 100% of the rent from the data center operator for its use of the building. Within the joint venture itself, the grid connection agreements and power infrastructure, such as the private wire networks and battery storage are held. And they will be jointly owned by Big Box and EDF. The data center operator will make recurring payments for access to this infrastructure from which both EDF and Tritax Big Box will each receive their respective 50% share. And finally, while there's no legal obligation to do so, it is probable that the data center operator will choose to procure its power through EDF. Their fleet of low carbon generation offers a compelling prospect to the operator for managing their carbon footprint. This will provide EDF with a single large customer for its power, helping to generate revenues for them. And this aligns with the interest of Big Box, securing a large rental income stream at a very attractive yield on cost and EDF securing infrastructure revenue and potentially securing energy sales. And we've already built a substantial pipeline. Working with EDF and applying the criteria I've laid out earlier on selection, we have established grid connection agreements that now total in excess of 1 gigawatt. And it's important to note that this pipeline is focused on the London availability zones, as highlighted by the orange shading on this map. We believe this area offers the best prospect for data centers and critically, with lower levels of latency on offer provides the best possible hedge against any future technological disruption in established areas. On the basis of our pipeline, we would anticipate most of the data center capacity will be used for cloud services or the actual user-facing implementation of AI models known as inference. We continue to add to this pipeline, and Tritax Big Box has further invested approximately GBP 100 million this year in securing additional power capacity with attractive and firm contracted delivery dates and land. For nearer-term grid connection agreements, we have now also secured the necessary land. And as a reminder, under the terms of the agreement between Tritax Management and Tritax Big Box, Tritax Big Box benefits from the first right of refusal on all data center opportunities Tritax Management identifies. It's the scale and timing and location of this pipeline that really sets it apart and creates a significant opportunity for Tritax Big Box. So I've outlined how we think about locations, how we go about securing power, and now I'll turn to our business model. Based on a number of factors, we have a preference for what is called the powered shell model of data center delivery. In a powered shell, we are responsible for delivering the land, the building and the power connection. Our clients will lease these assets from us and will be responsible for all the internal fit-out required to make it an effective data center. We've actually optimized this model a little bit further into what we're calling the Tritax powered shell, where we provide additional power infrastructure to enhance the quantum and resilience of the power. So for example, at Manor Farm, we will provide grid-scale battery storage and a plug-and-play power solution to the building. We believe the Tritax powered shell model strikes the optimal balance between risk and return and is well aligned with our experience in logistics, particularly our larger, multi-deck and very power-intensive buildings. The Tritax powered shell also builds on the value EDF brings, including the necessary power infrastructure, extensive experience at delivering large-scale infrastructure projects and clearly, a very credible name with large-scale and energy-intensive customers. We've walked through the strategy business model. So let's now take a look at our first live project. The Manor Farm scheme located within the Slough and Hayes availability zones will deliver up to 147 megawatts of data center on a 74-acre site adjacent to Heathrow Airport. We've selected this site due to its location between 2 transmission connections, a unique benefit in terms of resilience, but most critically, bringing first power on in 2027. The last capacity that can be delivered using existing infrastructure. The site also benefits from being bounded by the fiber routes linking London to Slough and then on to the United States. The project will also deliver a large-scale utility battery storage system, providing resilience to the grid, limiting the impact of the data center on the area. Planning has been submitted, and we're expecting a decision by the end of this calendar year with work starting on site in the new year. And I'll briefly hand you over to Ollie to give you a short tour of the site that I'm sure many of you will have seen it from the air as you depart Heathrow. [Presentation]

I hope you agree with me that that's really exciting. So now to the Manor Farm program itself, and I've shown here an indicative time line. In order to manage risk, commencement of the build phase of the data center projects will be conditional on both planning and a pre-let to a data center operator. This will allow us to work with the incoming tenant to tailor the data center to their specific requirements, reducing construction risk and enhancing the quality and longevity of the facility. As you can see in the graphic on the screen, the capital commitment in 2025 is GBP 80 million, including the purchase of the site for GBP 70 million. Capital spend during 2026 is then subject to planning and pre-let totaling GBP 120 million, and this is the bulk of the build cost, followed by GBP 65 million of expenditure in 2027, completing the Phase 1 build. We expect practical completion in the second half of 2027, at which point we will start to receive income. At this point, there is then a profit share payment totaling GBP 100 million. The funding outlined here is through internal resources and capital rotation with a target yield on cost for Phase 1 of 9.3% on a total commitment of GBP 365 million. So just to zoom in and summarize progress to date, planning is with the inspector and we're progressing as expected. We've had strong occupational interest from operators, and we have NDA signed with all the major hyperscalers and operators with their due diligence ongoing, which we aim to conclude with a pre-let conditional on planning by the end of this year. And looking forward, we're pleased to reveal today that Project 2 is already progressing and following a similar setup to Manor Farm. It's a 125-megawatt project situated in the broader London availability zone, scheduled for power on in 2028 and targeting a 10% to 11% yield on cost. At this target yield, the GBP 230 million of CapEx investment will generate potential rental income of GBP 23 million to GBP 25 million per annum. And as we've Manor Farm, construction will be subject to planning and a pre-let. So time for me to summarize before handing you over to Frankie to pull together the broader returns outlook across the group that you see on this slide. Our Power First strategy, plus our unique joint venture partnership with EDF is enabling us to deliver exceptional projects in time scales that are otherwise unachievable due to the power constraints in the U.K. market. Coupling this with strong occupational market and operator enthusiasm for the projects that we're bringing forward, we're targeting exceptional risk-adjusted returns for shareholders of 9% to 11% yield on cost, corresponding to the near-term rental income of GBP 58 million per annum shown here in the middle of the chart. And with that, over to Frankie.

Okay. So thank you, Tim, and good afternoon, everybody. I'm now going to take a few minutes to pull some of the good things coming out of those earlier sessions together and highlight how everything we've been hearing about translates into financial performance for the company and how we aim to deliver those superior risk-adjusted returns for our shareholders. It starts, as you see here, top left, with that clear set of growth drivers within the business, which Colin, Andrew and Tim have run you through. To aid in the delivery of these opportunities, we have built a very strong balance sheet to support us with a prudent capital structure, good access to the debt capital markets and corporate facilities, which provide us with a lot of financial flexibility in terms of our funding needs. In addition to this, we have multiple other routes to funding that provide us with optionality around the delivery of our strategy, and I'll come on to talk about a few of those in a moment. We're combining this to deliver superior risk-adjusted returns as demonstrated by the 9% total accounting return we delivered in the last financial year, being right at the top end of our peer group. And as Colin outlined in his introduction, the potential to deliver sector-leading earnings growth with an aspiration to grow adjusted earnings by 50% by the end of the financial year 2030. Throughout the presentation, you will have heard how the team focuses on maximizing the opportunity whilst consistently seeking ways to minimize risk, whether that's through the controlling of land in a capital-efficient manner to aligning land drawdowns to known occupier interest and operating with an appropriate mix of pre-let and speculative development. And in the case of data centers, focusing on the pre-let powered shell model. We are very conscious of our risk tolerances and ensure that as much risk as possible has been mitigated prior to committing significant capital into these respective opportunities. And this is a theme that spans all of our capital allocation decisions and is indeed the way we analyze the performance of our investment portfolio, viewing things through a risk-adjusted lens when selecting assets for either investment or disposal. As Andrew and Tim have highlighted, we're also deploying capital with precision, ensuring our capital deployment is appropriate for the local market conditions. The table on the left illustrates how we think of this in practice across each of our 3 growth drivers, showing the combination of capital intensity and risk at the point of that capital deployment. And the right-hand chart shows that whilst development yields across our pipeline sit at a healthy premium to both our existing portfolio and our underlying cost of debt with logistics at 6% to 8% and data centers now at 9% to 11% yield on cost. The arrows also show that these developments will generate strong levels of capital performance. And this is in the context of prime yields for both logistics and data centers sitting at around 5.25% today, providing us with a real kicker to our total return performance. Therefore, on a stand-alone basis, the returns that we can deliver from this pipeline are very attractive. But given our focus around the risk side of the equation, these opportunities are particularly compelling when viewing them on a risk-adjusted basis. Further to Andrew's comments on development management agreement income. DMA income forms part of our overall return profile. In response to questions from investors, here, I set out some further detail on both disclosure and guidance. DMA income principally relates to our freehold land sales where we undertake subsequent development for clients. It's important to note that DMA income is very capital light, typically requiring limited amounts of working capital only. As the bottom left-hand chart shows, we recognize the revenue side of DMA agreements, i.e., what we are billing our clients within other operating income, which is offset by the associated construction cost of the development as disclosed within other operating costs. The net position being the DMA income for the period, in this case, the GBP 23 million net position recognized for the financial year 2024. Given its nature, however, DMA income does fluctuate and the chart on the bottom right shows this has been the case for the last few years. So the policy we have consistently adopted is to guide to between GBP 3 million and GBP 5 million per annum to avoid introducing too much volatility into our recurring adjusted earnings measure. With anything above this level, we deem that to be additional DMA income. And this additional DMA income is excluded from the adjusted earnings measure from which we declare our dividends. And it's worth highlighting that this is real cash profit, and therefore, any additional DMA is reinvested back into the business, helping to enhance returns further and in effect, can act as a supportive funding tool. Turning now to our strong balance sheet, and you will recognize this slide from our full year results. Its strength is reflected by our low loan-to-value, a well-staggered range of maturity dates and with most of our debt being either fixed or benefiting from interest rate caps. As reported last week, we have successfully refinanced our GBP 300 million RCF facility, which was due to mature in mid-2026. We've now extended this out to 2030 and possibly 2032, subject to certain extension options. We've also increased its size to GBP 400 million, introduced new banking relationships whilst maintaining the same terms and conditions. With this debt profile, we are well insulated from the increases in debt capital costs that we've seen in the marketplace. And we have significant liquidity available to help us fund our strategy. As we illustrate on the right-hand chart, the pace of our rental income growth from reversion capture and the letting of vacant space far outpaces the likely increase in interest expense from any refinancings falling due over the period highlighted. We currently have a Baa1 rating from Moody's with a positive outlook and see the company's natural trajectory as one moving into the A credit ratings over the short to medium term. All of this gives us further confidence in our ability to drive future earnings growth for the company. And as I said earlier, we have a number of funding options at our disposal. You can see here the 4 main complementary levers that we've combined very effectively to executing our strategy and optimize returns for our shareholders. Over the last 2.5 years, we've realized proceeds from asset disposals totaling over GBP 700 million. This demonstrates both the quality and the liquidity of our asset base, realizing at or above most recent book values. We have recycled that capital to fund our growth opportunities at more accretive returns. And as you'll know, we are about half of the way through selling the GBP 475 million of UKCM's nonstrategic assets. And again, we will recycle this capital into the growth opportunities that we set out. In the past, we have also selectively raised equity when appropriate or more recently used our shares to finance the acquisition of UKCM. I've previously highlighted the flexibility within the balance sheet, and we currently have access to over GBP 550 million of available liquidity. And finally, whilst it's not a route we've actively pursued to date, we could potentially partner with third-party capital to support us with our financing. These levers are the funding engine of our strategy, and we carefully evaluate which combination of these we use at any point in time with our objective of driving shareholder returns whilst carefully controlling the financial risk. And so to conclude with some forward-looking guidance. We are continuing to invest to drive growth. In fact, as you will see here, we're increasing our guidance around certain lines of investment. We are maintaining our GBP 200 million to GBP 250 million guidance for logistics development for this year and out into the longer term. We're increasing our data center CapEx by GBP 100 million for the financial year 2025, anticipating GBP 200 million in total for the current year. The additional GBP 100 million is being used to secure further early deliverable power in the London availability zone, as Tim mentioned, along with as revealed today, securing the site for our Project 2. And as we build visibility on our data center pipeline, we're also increasing our data center yield on cost guidance to 9% to 11% from previously the 8% to 10%. We're also increasing our longer-term disposal guidance to between GBP 250 million and GBP 350 million per annum to support with this capital recycling program. And as a reminder that this year, we're guiding to between GBP 350 million and GBP 450 million of disposals, given we still have the balance of the UKCM nonstrategic assets to work our way through. Any delta in funding requirement can be supported by our balance sheet. Remembering, we will be able to apply debt to both the capital profit generated from developments plus the capital profit generated from rental growth without altering our broader loan-to-value position. And as we announced last week, with 2024 being the reference point, based on our 3 growth drivers, a self-funded business plan, we have the potential to grow adjusted earnings by 50% by the end of the financial year 2030. So to conclude from me, our strategy means that we are exceptionally well placed in sectors that offer structurally supported growth. And we have 3 very clear growth drivers, a strong balance sheet and a range of funding options to continue to provide superior risk-adjusted returns for our shareholders. And with that, I'll invite the panelists to come forward for the second Q&A session.

Great. Should we begin with some questions from the room. If you put your hand up, we'll get a microphone to you to ask away.

Matt Norris from Gravis. Just looking at data centers and rental growth, what's the trends there? What are we seeing? And looking further out into the sort of 3- to 5-year period, what sort of rental growth can we expect?

So as we said in the slide, so within the sort of 3- to 5-year period, we're not in a position to be able to necessarily talk to total rental growth yet. We haven't disclosed, obviously, the projects that we have in that pipeline, but the near-term 2 projects delivering in 2027 and 2029, will add GBP 54 million of rental income to the bottom line.

I mean, I think the general principle, Matt, as you heard earlier, when you're looking at quite restricted land areas where DC operators want to be against this acute shortage of power is only going to lead to one conclusion, which is that we would expect rental growth to outstrip the underlying growth that we see in the logistics market, which we already see as being pretty attractive. So it should be superior levels of rental growth in the DC market.

And contractually, what's in the leases? What should we expect? Is it annual rental bumps? Is the 5-year reviews?

So a lot of this is to be determined with the data center operators. Obviously, data center leases are quite different than logistics leases, but they do follow the same principles. So we would expect indexation in those leases. We would expect and we'll be fighting for market-to-market style rent reviews as part of those leases. But we enter into these negotiations with an open mind to secure the best deal for Tritax shareholders.

A question on the webcast about the spec for Manor Farm, if you can find any further details on the PUE certifications -- a lot of acronyms, OCP ready NVIDIA DGX ready, ISO accreditations, et cetera, et cetera. Tim, probably definitely one for you on that one.

Yes, sure. So probably not to necessarily deal into a list of certifications that we're anticipating for the design. But what we will say is that the facility that we're building for Manor Farm is designed from the ground up to be flexible for both cloud services and AI inference, I think, being kind of referenced there. We don't anticipate necessarily the ultimate occupational use of that facility to be AI focused. We very much see that as a cloud services focused/hyperscale deployment of those cloud service compute services. The PUE is designed at the moment to be 1.48. So that is kind of an average performance facility. We've done that on the basis that we can get the most flexible footprint from a planning perspective. And what we'll do is work with the incoming tenant during that sort of detailed design phase to hone and refine the building for their specific compute requirements. And that's the only way that you can get that PUE figure lower. It's absolutely key to the delivery strategy that we look to optimize that facility as much as possible from an ESG perspective and from delivering the maximum amount of IT operation that we can out of the facility.

I just get clarity on that 1 gigawatt question sort of when it comes through, what's contracted, a lot of this stuff seems to be kind of relatively a few years out. So if you -- a bit more understanding and apologies if I missed it as well.

Yes. So the 1 gigawatt pipeline, so all the connection applications are in existence. So they are there, they're real, they're tangible, and they have defined delivery dates with construction programs with the networks that pose limited challenges to delivery. It isn't the case everywhere. So we're quite proud of that. The sort of longer-dated nature of the time line compared to, say, a logistics development is just by the nature of what those assets look like. Power delivery dates tend to have a relatively long lead time due to the amount of infrastructure work that needs to be completed. A lot of that is actually outside of the kind of direct realms of what we necessarily deliver as the developer and a lot of that is delivered within the network. So lead times tend to be within that sort of 2- to 3-year time scale. In the sort of data center infrastructure world, they would be considered to be incredibly near-term opportunities.

And just sort of with a bit of blue sky thinking, what would be the earliest date you think that, that 1 gigawatt would be 100% drawn down in action?

So it becomes a function of delivering the right product at the right time for the users. I think our current assumption is that the earliest we would say that 1 gigawatt pipeline would be in the 2030 sort of time scale, but we can phase that in slightly longer-term time scales.

Some of them from the web chat. We heard -- end of Q1, we heard Microsoft had canceled multiple U.S. data center leases and pulled back from other pending leases raising concerns about overcapacity. Any comments on that, Tim?

Yes. So I was out at the data center conference in Cannes a couple of weeks ago. And I think the kind of the messaging from all of the data center operators and everybody in the industry is that this is not off the gas. It's not a slowing down of demand. This is about rebalancing of portfolios as you would within any other sort of development portfolio. I think the kind of the 10 leases that are off talked about were probably lower quality or less compliant projects within those pipelines that Microsoft and/or any other sort of occupier has better opportunities to take forward, and that's why we see those being left on the shelf, if you like. We don't anticipate that to be impactful on our portfolio at all. Our portfolio is very much more within the core zone rather than being in the sort of periphery of discretionary projects.

Sam King from BNP Exane. Just one question, please, on development economics and capital profitability. So looking at Slide 65, prime yields at 5.25% versus your increased yield on cost guidance of 9% to 11%. does that prime yield specifically relate to just the DC building and exclude the power infrastructure investments that you described under the Big Box powered shell model? Because presumably, the powered infrastructure is high yield versus the data center building itself. So should we think about the blended profit on cost of the entire project being slightly lower than what that 5.25% suggests relative to the overall yield on cost?

It's a good question. You're right. So the 5.25% is indication of the data center yield attached to that and the lease. The infrastructure piece, typically, you'd see something sort of higher single digits for that element. So blended, yes, it will be low 6s as a blend for the Manor Farm project. It's a good question.

Andrew Saunders from Shore Capital. I wonder if you can just talk us through what your take is on the delivery model that tenants prefer now for data centers. And I think where I'm coming from with this is you've been quite clear today about your powered shell with the power network provided. But a year ago, SEGRO did a Capital Markets Day where they told us that powered shells were the way forward, that was their preferred delivery model, yet we heard from in February this year that they've completely changed tack to go down the fully fitted route. And the reason given for that was that, that's what tenants want. So therefore, they're responding to that with the change in the model. So I'm just interested to know what your take is on which delivery model tenants actually want? And is there a risk that you might miss out by not going fully fitted?

So I think what we hear from tenants is that they don't want us to fully fit their data center. The kind of demarcation between landlord and operational use or the operational element of the data center is a really key differentiation for them. They want free and easy enjoyment and they want to operate their data center as an operational facility, and we want to provide them the kind of heavy bare metal by which they can do that. I think there are various reasons, and we can all kind of speculate why SEGRO might want to go down that route. I think that's not what we're here to necessarily comment on. But from a Big Box perspective, we don't think it is appropriate for Big Box shareholders at this venture or this juncture to be investing or involved with the operation of a data center. It's a very, very different prospect than leasing assets to blue-chip companies.

I think the other thing, Andrew, is, look, if you have several projects where you have firm power delivery, then you can pursue the powered shell model, okay? If you're looking at a horizon where you're struggling to get hold of the power, then you can't pursue that model as easily. From the conversations we've had with world-leading occupiers, as Tim mentioned earlier, there's a very, very clear and aligned desire for the model that we're currently pursuing.

Just turning to the webcast again. Tim, are there any constraints in accessing fiber capacity? And is there any value in having a JV with a fiber operator?

So the sites that we're looking at, no, currently, we don't have restrictions in fiber capacity. Those networks are pretty modern. What often comes down to being the constraint is the optical transmitter that you actually put on the end of the fiber. The glass is for all intents and purposes, the glass fiber that's there, and there is sufficient dock capacity. We do work closely with the fiber network providers, all of which or all of whom will be represented at Manor Farm, providing multiple points of access. So there's lots of capacity in these networks. We work with them early doors. If any additional capacity does need to be developed and deployed, that will be in line with the rest of the construction program in partnership with those operators. From a joint venture perspective, again, that's not something that we necessarily have a huge amount of expertise in-house in to manage. So what we would say is we will leave that to the networks to deploy that for the data center operators to the data center operator specifications.

Scott, there's a question, just the second row.

Callum Marley from Kolytics. Tim, I think you mentioned how existing distribution infrastructure remains the biggest headwind to supply. But how does the supply picture look over the next 5 years, say, following the grid connection reforms that are currently going on? Does that change anything?

So look, grid reform is a very, very wide-ranging activity within the networks and generators, et cetera. So it's changing quite a lot of the underlying legislation by which generation projects are managed within the network. The reality for the areas that we're talking about deploying data centers in, so particularly that sort of London and Southeast area, the constraints are physical in the network. So there just isn't enough copper between the sites. There isn't enough overhead line capacity to deliver that significant infrastructure. So you have to build more, and that's the reality of where we get to. Lead times on building this infrastructure is incredibly long. So we talk about some of the challenges that we have as a developer in the data center market. Those are mirrored and echoed by the infrastructure providers in terms of their lead times to deliver. So is there going to be a glut of capacity in 5 years related to the kind of talked about Iver B substation? No. The market isn't going to be a wash. That capacity has been long allocated to credible projects, not just data centers, but to growth in demand relating to heat pumps, electric vehicles and electrification of other activities. So all of this is kind of happening at the same time. So network constraints are going to persist for a very long time. This is about rewiring of Britain, not just kind of adding on a few extra bits and pieces. And that's why we are pursuing this Power First strategy, which is all about securing that real tangible network capacity when it's available and then deploying that to London infrastructure.

Tom Musson at Berenberg again. Sorry, just a quick one. Just how are you expecting the 50% adjusted earnings growth to translate into per share growth, just given the share issuance assumptions, the payaway assumptions we've got to make at Manor Farm?

So the context around that is obviously, it's a 2030 medium- to longer-term piece of guidance. It's an absolute piece of guidance. But behind that, obviously sits quite a detailed business plan and financial model, which assumes a self-funded model. So I think one can deduce that the absolute converts onto the per share basis based on that today's business plan.

A couple of questions again from the webcast. So probably one for you, Tim. With lighter models like DeepSeek, are we overestimating the power needed for inference?

I mean that's a very good question. What I would probably say as a bit of a personal opinion about the development of this market. There's a lot of talk about growth in the AI market. But as kind of Kevin refers to, the actual growth is in cloud services and wholesale realistically in the market. So while these are very rapidly growing markets, they really aren't coming from the same sort of size and depth that the cloud services markets are coming from. With DeepSeek, in particular, although that model is claimed to be lighter in terms of the way it's trained, and there are some critics who may say that, that isn't quite necessarily the case. All this does is to drive that technology into users' hands. So the more available these technologies are, the more common they will become in our day-to-day lives. And that is fundamentally what's driving the demand, particularly in our portfolio where we're talking about delivering those cloud service peripheral AI technologies, the inference element, which is the bit that you interact with, that is effectively a cloud service. So the more people using these technologies, then the demand for our portfolio and the style of the facilities that we're looking to deploy will grow in line with that.

Great. A flurry of questions coming through on the web chat. What kind of rent-free periods do you expect to have to give to the tenant at Manor Farm given it will take them a year to fill out once you've handed over?

So our current position is that we don't want to offer a rent-free period to operators, and that reflects the highly prime and very priced nature of the assets at Manor Farm in terms of the scarcity of the resources. That will come down to a contractual negotiation with an operator.

And to get another set of question, have you had any early indications of the range of rack densities, potential occupiers of Manor Farm might be targeting?

Quite reasonable range. So from 4 kilowatts upwards. We're not looking at rack densities associated with AI technologies. That's not what we're envisaging, although the building is capable of taking those large DGX racks as one of the previous questions refers to.

Okay. And then a further question relating to the yield on cost. Does the new opportunity announced today have a similar return profile to Manor Farm, i.e., yield on cost disclosed here at 10% to 11% and exit yield around, say, 6% to 6.5%. Is the total development cost per megawatt also similar to Manor Farm?

I think from a yield perspective, both of those statements are probably fair, 10% to 11% and something in around 6% blended. Second, Ian?

Total development cost per megawatt.

So Project 2 has a slightly lower total development cost per megawatt, and that's just on the basis of the amount of infrastructure that we need to deliver. And this is just a function of where our grid connections are compared to our site and how we go about connecting the infrastructure.

Then, Frankie, probably one for you. What are your assumptions for your marginal cost of debt for the 50% adjusted earnings growth aspiration?

We are effectively applying the forward curve to those facilities that are up for refinance, and we are effectively bringing that refinancing point in 12 months prior to its effective maturity date in the assumptions behind that chart.

Great. And again, sorry, Tim, probably one for you. Are your assumptions on rental terms based on pounds per kilowatt or on pounds per square foot? .

So we currently work off pounds per kilowatt IT, though we can readily convert between that pounds per kilowatt IT and pounds per square foot. So it's just a simple conversion between the 2.

Again, sorry, Tim, one for you again. Are you involved in cooling infrastructure? Or is that solely with the tenants?

Cooling infrastructure is for the operator to deliver.

Okay. Any other further questions from the room? Okay. I think we might hand back to Colin to wrap up.

Well, thanks, everyone, in the room and on the webcast for your questions. Well to conclude, we've outlined here how our investment case offers investors attractive returns supported by our 3 growth drivers. And this growth is underpinned by quality and efficiency. Quality in our modern sustainable assets, supportive long-term markets and our world-leading clients and efficiency in our lean and agile structure, triple net leases and our strong balance sheet. These combine to drive earnings growth, dividend progression and the potential for further capital appreciation. And I strongly believe that we have all of the necessary attributes to be successful in the short, medium and longer term. And I hope that today's session leaves you with the same confidence. And as I said earlier, our team is unfortunately unable to join you, but refreshments will now be served on the terrace. I believe the sun is shining, and that concludes today's seminar. Thank you so much for joining us.