Mesa de Real Estate en el Andorra Economic Forum 2024 - Ismael Clemente y Juanlu Cruz

Well, it was a bit of a succession of coincidences. I was working at a multinational, at an investment bank. I was doing management, doing principal investment. And in 2012, coinciding with practically the bottom of the 2008 crisis — 2007 to 2014 was especially virulent in Spain — my company was a bit embarrassed about having a headquarters in Spain and wanted me to make a series of adjustments to my staff. We were far from unprofitable; we had a cost-income ratio of 30%, so we were a super profitable unit. But out of a sense of embarrassment, they wanted to act, wanted to give the appearance of doing things, like politicians, right? So they asked me to make a series of adjustments and I refused. I went home and ended up unemployed. Since I have six kids, I had to find a job, so I had to set up a business together with my colleagues who also left with me. And since what we knew how to do was investment management, that's what we dedicated ourselves to. We were lucky that Blackstone chose us as a management platform for Spain. After a couple of magnificent years, Blackstone shifted from doing REOs, from buying assets, to doing NPLs, to buying loans. We didn't really agree with that strategic shift, so we decided to organize on our own and we had to set up a SOCIMI, and that's where everything started.

Well, a SOCIMI is an American invention from the 1960s. After World War II, housing prices in particular, but real estate prices in general in the United States, had risen at a compounded rate of 7% and had tripled. So the American government started thinking about creating some kind of vehicle to channel savings so that people wouldn't buy things on their own but would have a bit of professional management. So they decided to create a figure that's halfway between a corporation and a fund — I say halfway because it has a legal regime whereby it doesn't pay corporate income tax, so as not to be penalized compared to funds, because otherwise people would have kept putting money into funds rather than into a corporation. But in exchange, it's obligated to distribute dividends upward every year, and those dividends are taxed above. So this political mantra that it doesn't pay taxes and so on is simply a falsehood. They wanted it to be a corporation that had the capacity to act in the market, that could develop properties, manage them, operate them long-term, that didn't have a speculative vocation — that didn't buy and sell properties but held them in its portfolio for a long time — and in exchange, from the rents it obtained each year, carefully minimizing operating expenses, it would distribute dividends upward. And with that dividend, a schoolteacher from Wisconsin goes on vacation to Florida every year thanks to the dividend. In the United States, 153 million people own shares in REITs. It's an absolutely popular vehicle, a normal way of saving and investing for pension plans and funds. In Europe, it's much less well known, and in fact, we're constantly subject to public debate and scrutiny, when in reality this is a vehicle that already exists in 69 OECD countries — in all the good ones, that is: Japan, South Korea, Australia, the United Kingdom, France, Italy. So it already exists in a lot of countries, and well, that's what a SOCIMI is.

Well, there are two types of SOCIMIs. Those that trade on primary markets, the continuous market, or in this case, they may or may not be on the selective index. Those SOCIMIs clearly have a vocation of share dispersion and maximum liquidity. You want your liquidity to be maximal, so that a person can buy your share at noon, sell it at three, buy it again at five, and sell it the next day at eleven. That's the idea. Then there are a whole host of SOCIMIs on the secondary market, on parallel alternative markets, that don't have immediate liquidity and don't have a vocation for it. They're vehicles through which a family office or an investment fund makes an investment, obviously distributing dividends upward and complying with the regime, but they have no vocation for minority investment. So it's a completely separate reality, and there are over a hundred of those — between MAB, portfolio, Euronext Paris, there are a lot of Spanish SOCIMIs listed on alternative markets. Well, that's how it works. But in our case, the vocation of dispersion was complete, and what we did was, with that idea from the beginning, do a kind of experiment — like a final thesis project. Let's set up one of these things and it's going to be 100% free float. So initially we thought of creating an externally managed vehicle of 3.4 billion in volume, and when we started the fundraising, we went to do pilot fishing, mainly in the United States. We saw that we were gaining traction. It was 2014 and Spain looked like a wasteland — it was like a Mad Max movie. You'd drive down the road and there was nothing but smoking ashes on either side. And well, when we went to the United States, we suddenly saw that there was probably more water in the pool. The banks started telling us there was enough tension to get 600, 700, then 900 million. And from there, we switched from external management to internal management, because at certain volumes, investors don't want you to have an external manager — they want you to be an employee. And at a certain point, when we saw we could reach about 1.3 billion, what we did was, to avoid cash drag — so the money wouldn't sit in a bank account for a long time causing a drag on shareholder returns — we started accumulating a series of investment opportunities to be very quick as soon as we finished fundraising to invest those funds and have cash flow from the beginning.

We opted for a fairly safe method, which was to buy an investment we had previously made in our previous life at Deutsche, with a series of funds of which there were American and British funds that had already completed their investment cycle and wanted to leave. Besides, in 2014, for many people, the recovery of the Spanish market wasn't yet clear. For us it was, but then we bought that. And by buying a sale-and-leaseback portfolio, you ensure two things that are very necessary at the beginning of creating a company: a lot of cash flow and little need for management, because you're going to have to manage the creation of the company, so the less you have to manage the assets, the better. So it seemed like a magnificent initial deal. We went all in on that, and with that foundation, we built the rest of the building.

Correct.

Well, in our prospectus, in our IPO prospectus, we always made it clear that we wanted to do retail, offices, and logistics. The sale-and-leaseback thing was a bit circumstantial and a way to start, but we were clear that we wanted to work in the main segments of commercial real estate without getting into the perilous world of housing, which has a very political component, and the further from politics, the better. And then hotels — well, there are a lot of people, there are many specialists, so we could stay away. And I love hotels — as an asset class, it's probably my favorite — but one doesn't have to do what one likes, but also sometimes what one must do. So we wanted to do that. But what we did at the beginning was play the cycle in the following way: at first we set up small cells spread across the territory, buying properties throughout Spain and Portugal a bit later, assuming that assets you buy in such an early phase of the cycle, you'll be glad you bought them a bit later because they improve in yield. But second, we also quickly realized that that trail was going to fill up because other funds were going to arrive — mainly American opportunistic funds — and they were going to outbid us because they were bigger than us at that time. So we quickly moved to a second acquisition strategy, a bit unusual but that gave very good results, which was buying existing corporate platforms — buying century-old real estate companies that had succumbed to the crisis and were in very difficult situations. We were able to buy two, not because we liked being in the press, but because by buying those platforms, we accessed at once an enormous collection of assets below value. That enormous collection of assets is what allowed us to suddenly have logistics centers and offices and start sorting things out — this one goes here, this one goes there, this office with this, this center with that — and start creating what today are the divisions the company works with. We have a fully operational division to which the company merely serves as an umbrella, providing rating, financing capacity, and capital. But we have a shopping center team that competes with shopping center companies, an offices team that competes with office companies, and a logistics team that competes with logistics companies. The data center thing was a bit of a coincidence and the result of a reflection. We thought back at the end of 2018 that the expansive monetary policy that central banks had followed in a coordinated way for many years was going to end at some point in inflation, and that inflation brings higher interest rates. For a real estate company, it's absolutely essential to watch interest rates, because many investors treat you as a kind of bond proxy. So if interest rates go up, your share price goes down, because they make a syllogism. And that's even more so the more leveraged you are. So we realized that at some point rates were going to rise, we needed to radically reduce leverage, and we decided to sacrifice our foundational asset base — the BBVA bank branches — because it was an eminently analog asset, but with very long-term lease contracts hyper-indexed to European inflation. As soon as inflation appeared in the market, the price of that type of asset, by discounted cash flow, was going to skyrocket and be much higher than its underlying value as a real estate asset. So it was the moment to sell it. But if we sold, we needed an income replacement, and that's when we started thinking about data centers. In 2019 we started going to the United States to understand that business, and in 2020 we signed the technology transfer agreement with Edge, which has allowed us to have a different cooling technology.

A data center is basically a facility where, in an energy-efficient way — and I'll explain why it's energy-efficient — data is stored in bulk and transmission is facilitated between data centers and with end users. That's what a data center is. It's a kind of logistics warehouse where, on the Mecalux racks, instead of goods, you have data. And by the way, they're also called racks — the cabinets where that data is stored. Why do they exist? Because all of us make them exist. It amuses me to see in society a bunch of people who — I don't know why — have created a somewhat negative mantra about data centers, especially among radical left groups. The data centers and the radical left — I understand you're referring to the PSOE. Well, I hope the PSOE has more vision than this. Look, it's true that data centers consume a lot of energy, but it's the energy you generate with your device. If you go burn containers with your face covered because you're a super revolutionary guy, and then you make a TikTok with your friends about burning containers and send it to a group of 55 people who each forward it to another 55 people, you've just generated 284 gigabytes of data traffic, and you generated it yourself. It's a curious thing. Data centers consume as much energy as we demand through internet applications. So why are they growing exponentially? Because this is growing, because artificial intelligence usage is growing. AI has just arrived on smartphones — Samsung was the pioneer, they put it in around June, and now the rest of the manufacturers have been adding it. And today, AI applied to smartphones is barely 1% of data traffic. But this is going to grow exponentially. And by the way, as with almost all technologies, the gap between the first, second, and third worlds is going to close, so that in a few years we're going to see people in places like Djibouti — who today already operate drones from their phones — accessing artificial intelligence everywhere in the world. So what this causes is an exponential increase in the need for data storage and transmission, and that's why data centers exist.

You have three types of clients. Corporate clients, which is what we call Enterprise Traffic — a large company, let's say BBVA, which through its technology integrator, let's say Equinix, rents space in your data center. Equinix does the necessary adaptation for them to set up there and stores their data in your center. This is very lucrative traffic because the price per kilowatt is very high, and you also sell them a bunch of auxiliary services, interconnection, and so on. That sells at a high price and it's wonderful. The only thing is that it's absolutely minority traffic — meaning today, practically no large Spanish corporation exceeds one megawatt of storage needs. In the United States it's a different story; they're much more digitalized, but not here. So that's a very lucrative way to fill the center, but it takes forever to fill it. So each company has to choose how to position itself. The second major source of clientele for a data center is the hyperscalers — Tier One, there are four or five, and then there are some Tier Two. These are wonderful because they take blocks of three or five megawatts, but in exchange, since they're an oligopoly, the contract basically has a dotted line at the bottom where you have to sign — so your negotiating power is limited. And the third major wave, which has been very recent for about a year, are the artificial intelligence operators. These are the ones taking blocks of megawatts of computing capacity like there's no tomorrow. And there are many more of them because a lot of companies have been born under the AI boom, capitalized by venture capital, and right now they're sitting on a pile of cash with a need to find space to store their computing needs. Their business model — many of them, especially those doing AI as a service — is very simple: they buy from you a block of computing capacity per year with a lease contract, and they resell it to you by the minute. So what they do is essentially a kind of wholesale-to-retail trading, and that trading is brutal. So much so that you, as the provider of critical data center infrastructure, are like the leather manufacturer from Ubrique who sells leather to Loewe — you sell one square meter of leather for 50 euros, and they make three handbags worth 5,000 each with the leather you sold them.

Today, yes. Today, yes. But well, today, yes, just as logistics was in 2011. We bought the first logistics warehouses at MERLIN in 2014 — I remember one perfectly at a 10.75% yield, at 9.25%. That was normal at that time. All sectors, when they end up being crowded, yields compress. So right now, you have to know that the gross-to-net in this sector is quite severe — you have a gross-to-net of about 70%. So the gross yield you get, which might be 12% or 14%, doesn't matter because what matters is the net yield, and the net yield is going to be between about 8.5% and 10%, which is above what you normally get in real estate, but not that far above what you were getting until recently in logistics. So that's going to compress, obviously. Those who had a first-mover advantage will have the opportunity to enjoy oversized cash flows for a long time, which is great. But the market rationalizes everything quickly and yields will compress over the years.

Very good question. We're a bit everywhere, understanding what last-mile means in Spain. In New York or London, to do last-mile — which is normally from the break-bulk point to the delivery point — you need less than half an hour. To do that in London or New York, you have no choice but to put infill warehouses inside the city — small, inefficient warehouses from a logistics standpoint because they usually have columns, and you operate uncomfortably in them. But you have no choice but to look for that inside the city to do last-mile. However, given the size of our cities, both Madrid and Barcelona can perfectly serve last-mile from the first isochrone. So each of our warehouses located in the first isochrone is serving last-mile. In Barcelona, Parc Logístic, Zona Franca is serving last-mile. And in Madrid, everything we have in Coslada, Getafe, San Fernando is serving last-mile. Maybe if you go to Cabanillas del Campo, not quite, but almost. Our cities aren't that big or that congested with traffic, despite the fact that we all have an image that our city is always congested. Our city is congested if you compare it to my hometown, but if you compare it to Delhi, it's very uncongested. If you compare it to New York, it's very uncongested. So that's last-mile. We have a bit of everything because in the end we operate 2.2 million square meters, plus what we have under construction, so about 2.8 million. We're by far the largest logistics operator in the market. We have everything: from triple-XL warehouses — enormous ones that are national distribution centers for companies, typically between 50,000 and 100,000 square meters — to regulation warehouses that are much smaller, cross-docking, cold storage — positive cold, negative cold — we have ADR for chemical products, and we have last-mile, which is cross-docking much closer to the city. So we work in pretty much all segments.

Well, in our case, the main threats, unfortunately, because we're a Spanish company, are always in the regulatory framework. Everything else is going very well. Right now, everything is fairly aligned from a strategic standpoint. What we'd like to do is gradually balance the company's income sources — shifting from a completely analog income source, as we've historically had, to an income source that's partially or even mostly digital. And for that, the data center division was created. If you look, you might think that technology companies are indirectly also a digital source, but in our office portfolio, we have about 12 buildings with a lot of companies, but most are industrial companies — automotive, pharma. There's also banking, insurance, aerospace industry, electricity — it's very diversified. Our main characteristic is that 95% of our clients, in the CNAE classification, are large companies, not SMEs, and 70-77% use their space with us as their headquarters in Spain and/or Portugal. So in that sense, we're very well positioned, but it's an analog source. What do we want? More digital sources. How do we start achieving it? Through online logistics. Online logistics depends on digital commerce and therefore gives us a balancing factor with shopping centers, which are physical commerce. So if one side fails, you balance with the other. So we started first with online logistics, and now we've added data centers. Our strategic ambition is that by 2030, we can have an absolute balance — 50-50 or even a majority in favor of digital income sources versus analog income sources.