Latest / The Promote Podcast / Deconstructing Data Center Megadeals with DigitalBridge's Marc Ganzi
Transcript
- Will Krasne (00:00) Remember the Gene Hackman, Will Smith classic, Enemy of the State? Hiten Samtani (00:04) 100 % Gene Hackman (00:05) the building why because you made a phone call Will Krasne (00:09) When chatting with this week's guest, I was reminded of Gene Hackman's character in the film because like Brill, Marc Gansey was very early and very right about data in the cloud. Hiten Samtani (00:19) You're reaching there, but I'll allow it. Welcome back to the Promote Podcast, your insider guide to the money and mania of the CRE markets. I'm Hattensom T. Please don't leave out your middle name, your will, co-host of a Top 100 Business Podcast, Krasny. Will Krasne (00:30) and I'm Will Krasny. My parents had a lot of foresight naming me that. Hiten Samtani (00:39) Thank you guys for all the love so far. Keep it coming. We're both egomaniacs and we need the weekly validation. Will Krasne (00:45) More often than weekly is great because as Don Draper said, happiness is just a moment before you need more happiness. Hiten Samtani (00:53) This week, we catch up with one of the megawatt dynamos of the data center space. I got to know Marc Ganze in 2021. He was in the midst of an overhaul of Colony Capital, the former Tom Barrack joint. Now, Barrack's Colony was a mishmash of hotels, industrial, debt funds. Ganze's Colony, though, was a whole different beast. So Ganze saw an opportunity to put himself at the center of the digital infrastructure conversation. He later rebranded the company as Digital Bridge. and he went all in on data centers, cell towers, small cells, and fiber network properties. Then came the AI boom. Capital from all over flooded into data centers, as we talked about a couple weeks ago, Yeah. The billions of dollars from all over the world just seemed to be flowing into this one asset class. But Ganzi sees all these players are coming in as Johnny cumblateles. A shout out to our sponsor for this episode, MADDPROJECT. They're a New York City focused resi developer. You'll hear more about them in a bit. Well, look, I love this conversation. I think we both learned a lot, but it was also a challenging one. Marc's a good talker. The problem is he's a little too good. Will Krasne (02:00) Yeah, there are other podcasts where you can hear Marc Evangelize. We wanted to avoid a conversation purely about the merits of the space itself and get more into the nitty gritty of how you build this type of company. Hiten Samtani (02:13) we decided to zero in on two defining deals. The first one is Digital Bridges $11 billion take private of data center operator switch in 2022. The next one, a fascinating deal, Coreweave's metaphorical lockbox, which is a $7.5 billion debt financing that Ganzi helped structure last year. So those two deals kind of give you a sense of what he's up to and what the space is up to as well. All right, let's go to our conversation now with Marc Ganzi Marc, thank you so much for being with the Promote Podcast. I think we want to dive right into a couple of defining transactions for DigitalBridge. Let's start with the 11 billion take private of Switch. Marc Ganzi (02:44) Great to be here. So for a long time, I've watched from the sidelines, the evolution of switch from when it went public to, you know, how it ran as a public company. And it struck me that asset was just mispriced and, and Rob's a total pioneer. Hiten Samtani (03:09) He's referring to Rob Roy, the founder and CEO of Switch. Marc Ganzi (03:12) He was evangelizing, you know, data sovereignty, like a decade ago, way ahead of everybody else. And his view was that while the public cloud is innovative, it's accelerated growth, it's, it's, it's been a big driver for the economy. And it certainly has accelerated the digitization. There is a peril to public cloud, which is that your data is out there. And Rob was like, look, I've, I'm going to make a bed that there's going to be certain customers that really appreciate a bespoke. Experience where the security and the reliability and the connectivity are second to none. And he was right about that. There are a series of fortune 100 customers, including the U S government and other, big organizations that really want their data to be safe. And Rob is just obsessed with security. And he's the only data center operator in the world. That's never had one minute of downtime. That's a record. The second thing. that Rob has going for him is he has a really unique ecosystem of interconnection. And as you know, data centers are very reliant upon fiber connectivity and he owns essentially one of the most important interconnection junction boxes in the world, which is in Las Vegas. And so Rob had this really unique ecosystem of 14,000 different carriers cross-connecting in Vegas. Will Krasne (04:27) That point about location, I really, wanna double click on because in intervening, what, 18 months or something, Cyrus one got taken private, Corsair got taken private, obviously QTS got taken private. So was this one you were waiting for? And then secondly, on the location point, it seems like right now a lot of the data center, the big data center construction, like we talked about the OpenAI campus and Abilene. Abilene, yeah. Is taking place sort of in the middle of nowhere. And it seems like the market and even the debt markets are valuing. these assets irrespective of locations like Ashburn is the same as North Dakota. Hiten Samtani (05:01) mean, air rights in Manhattan are worth something. Air rights in suburban Florida aren't worth the same, right? So how is it that in the data center space, the market is treating those quite equally? Will Krasne (05:12) And was that part of the equation as well for square Marc Ganzi (05:15) So everyone sort of made the assumption like, my God, they're not a Nova and they're not in Santa Clara. So therefore they're not in May East, they're not in May West, so they missed. Well, what they had is they had May Vegas. It's as valuable as May East and May West from a connectivity perspective, but nobody really gives that particular asset its shine that it's worth. I think there's no doubt in my mind that the residual value of data centers, guys, in Santa Clara today is much higher than a data center in Dallas. Why? the residual value of land. you want to buy an acre in Santa Clara, it's like $15 million an acre in that industrial area. If you go buy an acre in South Dallas, it's like 700, 800,000 acre. But at the end of the day, whenever you underwrite a data center, you know at some point that the customer may leave, right? At some point. So there is will this inherent value in the land and you got to get comfortable with the land, the entitlements, the power, that basket of real property rights that you have. You got to put a value on it at the end because if it turns out that after 15 years of collecting investment grade cash flows, the customer leaves and goes somewhere else, you better have an alternative purpose for the real estate. Hiten Samtani (06:26) What alternative purposes exist in this space? mean, the lead times on such projects for new projects are massive. The capital requirements are immense, right? And the complexity and the regulatory hoops that you have to jump through take forever. So one of the broader questions we have about this space and which is why we kind of wanted to have you is how long can state of the art be state of the art? We're talking about building the future here. You talk about this a lot. What if you're kind of already in the past? Marc Ganzi (06:54) No one can sit you with a straight face and tell you their future proofing and data center. That just doesn't exist. I think there are certain things that we can do as an operator. There's a letter level of cooling we provide. There's a level of security we provide and redundancy ⁓ and a track record. Those are four things that are real critical selling features with a customer. The reality is the customer is going to change their equipment two or three times in that 15 years. They're going to change the server. They're going to change the GPU. They're going to make changes. And I have a pretty good view on this because for 32 years, we've been in the tower space and we've had customers sign 25 year contracts and they've changed their radios and antennas five times. We refer to that as the drop and swap. They drop the old gear, they swap in, put in the new gear and usually the rent goes up. Will Krasne (07:46) That has a different connotation in Pennsylvania. It's a way to avoid transfer taxes. Marc Ganzi (07:50) Yes, I'm familiar with this term. Now you have the tower one. So you have both in your pocket. Well, you can use them as you wish interchangeable. Will Krasne (07:56) Dual threat now, perfect. Hiten Samtani (08:01) This episode is brought to you by MaddProject, a New York City-focused developer with over two decades in the trenches. Antonia and her teams have delivered over a thousand residential units in the city, and the opportunity is clear. Now's the time to be building for-sale, roundup, and repositioned resi projects. They're seeing steady demand from high net worth buyers and a 6 % year-over-year increase in sellout prices. So they've launched a development fund to meet this demand, and it's open now to accredited investors. They also have deal-level syndication options for potential anchor investors or JV partners. And they can work with owners interested in converting or redeveloping their buildings. Hit Antonia up at Antonia@maddproject.com to learn more. That's Antonia at maddproject.com. Thank you. Back to the episode. So I want to understand again from a real estate perspective, you acquire this thing, what are you bringing to bear in terms of your real estate playbook? Will Krasne (08:58) You're going up against Brookfield, right? For the... Marc Ganzi (09:01) Yeah, Brookfield and we were going up against Equinix. Will Krasne (09:04) Yeah, so two companies that are larger, bigger war chests and the digital bridge. how do you... I know. So how does that come to bear? if it comes down to who can say the bigger number, the other two guys can say pretty big numbers. Marc Ganzi (09:10) I'm the under- I'm the underdog. Hiten Samtani (09:17) It's not gonna be you. Marc Ganzi (09:19) As the proxy reads, we were not the bigger number actually. We were the lower number, but at the end of the day, you know, management had more confidence that we could execute the growth plan than the other guys could. Now we fed that narrative. We convinced the management team at switch that we could double the size of their business in five years quickly. And the playbook was simple, which is one, we underwrote the land and the power. What Rob had that nobody else had was a power bank. So he had a six gigawatt power bank. Nobody had that. And he was only using like 800 megawatts when we bought him. Hiten Samtani (09:56) Have you seen Scarface? Yeah. There's a famous line in that. Marc Ganzi (10:00) famous line this country you gotta make the money and then you get that you get the power yeah and then then you get the Hiten Samtani (10:09) Exactly. So data centers, feel like the power is the first riddle you got to solve. Marc Ganzi (10:13) You got it. You got to get the power. You get the power. The power is critical and you have to have land, right? You got to have a land bank to grow into. Yeah. And look, these guys, the end of the day were sitting on 1600 acres of land. Nobody had 1600 acres of land for data centers. Nobody was thinking in large language models like Rob Roy was three years ago. Hiten Samtani (10:34) How do you underwrite on that point? How do you underwrite the regulatory risk? So Northern Virginia data center alley, there's now lot more hostility and backlash to this kind of thing. And I'd imagine this is happening in other big hubs, right? Marc Ganzi (10:47) We love the fact that Rob was not in zoning sensitive markets. He just wasn't. He was like, I don't want to be in Northern Virginia. I don't want to be in Santa Clara. And I was like, what? He's like, Marc, you have to understand AI is coming. I'm like, I agree. And he's like, you know, there's going to be like hundreds of gigawatts of power to be built. And I'm like, no, you're crazy. You're crazy. You know, we're essentially at 40 gigawatts today. The industry is going to like 120. He's like, you're wrong. The industry is going to 300. And what's gonna matter is that you have great connectivity, and you have a ton of land, and you have a ton of power. And we at Switch have. Hello? Hiten Samtani (11:28) we lose them? I think so. Okay, as we're interviewing one of the Gs of digital infra, our frickin' digital infra goes down. Will Krasne (11:40) Yeah, I think we really need to have an actual lockbox, not the core weave metaphorical lockbox for our Wi-Fi connections. Hiten Samtani (11:48) we were able to improvise and Marc was a great sport about it. Just a note for listeners, the audio from here on out is gonna sound a little bit different. You see him? Marc Ganzi (11:58) Not to be undone by technology. Will Krasne (12:01) I was gonna say if we were at a Switch data center this wouldn't have been problem. Marc Ganzi (12:05) Totally would not have happened. Hiten Samtani (12:07) We were talking about like kind of the regulatory risk of being in certain hubs, Northern Virginia. Marc Ganzi (12:12) That is the key here is, some people don't understand that you can go into a market like Northern Virginia and go to a market like Santa Clara and then the power companies say we're out of power and you're just done. But what we identified early is that large language models did not need to be in Northern Virginia and they didn't need to be in Santa Clara. They could be in places like Texas. They could be in Reno. They could be in Las Vegas. If you have strong fiber connectivity and low latency. You know, these large language models don't need to be at State and Maine. Hiten Samtani (12:43) is so interesting because you're saying this is the first real estate asset class that is untethered from location, location, location. Marc Ganzi (12:50) AI is going to happen in three phases like public cloud did. Public cloud, they first built their big high-powered nodes, which were not location intensive. Then they went to their first AZs, and then they went to their secondary AZs, and then they went to the edge. The way AI is playing out is very similar. First quarter of the football game is LLMs. Second quarter of the football game is generative AI, where the model starts thinking for itself and it starts making decisions. The third quarter of that football game is inference. where it thinks for itself and it starts doing. And that is very location intensive. That is super location intensive. And you do need to be next to the enterprise. You need to be next to that power grid. You need to be next to that consumer user. You need to be next to that government agency. And then the final is, know, inference to the edge. And inference to the edge is this, to your mobile device. And so the initial innings of cloud and AI played out exactly the same, which is they are not location intensive. But as you start building AI and now everything is about the edge, near edge, adjacent to the application, we just happen to be in year three of a 10 to 15 year build out. Will Krasne (14:02) So how hard is it then to go against conventional wisdom and convince the capital to come with you because the orders of magnitude of capital needed for one of these facilities is so large. mean, we were joking last week that the Abilene Data would be larger than a small cap public company. Hiten Samtani (14:23) going to say it was half the Hudson Yards capitalization exactly. Will made the point that was like the power demands from that thing were the same as the power demands in the Back to the Future thing. What was it called? Will Krasne (14:35) 1.21 gigawatts. Marc Ganzi (14:37) Point 21, sicko ass. Hiten Samtani (14:39) Yeah. You're also seeing kind of the biggest of big players in global capital flow in, right? You've got Apple, that he wants to be a massive player in this space. All the sovereign wealth funds are trying to see how they can get in, et cetera. So I wonder like, as a real estate guy, what the edge becomes, capital is not your edge, right? When all these kinds of people come in, capital is not the edge. Marc Ganzi (14:59) The edge for us is being able to sit with a guy like Rob and articulate very clearly how we're going to make him a lot of money. Rob, if you trust me, I will one, form capital. Two, I will raise debt capital like you've never raised before. We will securitize all of your data centers and we're going to do the first green wrapper data center securitization because you're 100 % renewable. And the agencies are going to love it. Investors are going love it. And you're going to price your debt cheaper than you've ever cheaped it. You better dumb for it. Three, we're gonna bring the public cloud guys to you. You've historically have said no to them. We're gonna get you to make peace with those guys, because they're big customers. They take a lot of space. And for us to double the size of your company, you know, from going from 287 million EBITDA to 560, 580 million EBITDA, we're gonna need the public cloud guys. By the way, we were wrong about that. And then last but not least, we're gonna get you to a spot where you're free cash flow positive and your business is self-funding. We're essentially two and a half years after closing the deal and we've quadrupled Rob's EBITDA. Hiten Samtani (16:08) I know you can't comment on this, I should just say for the record, there's a rumor that there's a 40 billion ish IPO on the works. I came out in Reuters a few months ago. I'm just saying it for the record. Marc Ganzi (16:17) Like, look, if you extrapolate that we've quadrupled the size of the business, more than quadruple the size of the business, you can get very comfortably to an even a number well north of a billion, three billion, four, and you can start running the numbers around DLR and Equinix is multiple and you get, you can get there. think that's how people are backing in to that number. Hiten Samtani (16:37) So Will, this was my favorite part of the conversation. Marc hits on a really powerful force in investing, a force that I think is more potent than most people give it credit for. FOMO, capital goes where capital already is. That's why every sovereign wealth fund is falling all over themselves to invest in data centers. That's why every big CRE player is forming a new JV to do the same. Will Krasne (16:45) yeah, FOMO. These way to not get fired is to invest behind the black stones already doing. Hiten Samtani (17:01) Yeah, you've talked about error cover before and I think this is the perfect kind of example of that again. Let's jump back in. Will Krasne (17:05) 100%. Hiten Samtani (17:09) There are a tremendous crop of new players in the space. Everyone from obviously Blue Owls on the Abilene deal to Related has just formed a data center at JV. A lot of like real estate operators obviously want to get in on this action. Marc Ganzi (17:22) Don't forget Barry. He'll be left out if you don't say Barry. Barry has FOMO's, you know. Will Krasne (17:27) As a former member of the Starboard Hotel team, I'm very sad that that allocation has gone from hotel to digital. Yeah. Hiten Samtani (17:33) This is a great, I mean, you're saying it in jest, but I think the data center space in general has a large FOMO component to it as well, which is why you're seeing this tremendous flood of capital coming in. What are some of the things that they, as real estate people, maybe not necessarily as data center people, might not be aware of or need to think about? Will Krasne (17:53) And just to hop on that, think you can talk about the second topic we really wanted to get into, which was the core way financing, because that one is where if you're just a real estate guy and not a operator, like what are you going to do with the collateral at the end of getting it? Marc Ganzi (18:06) Here I am with a bunch of GPUs, Will, what are we going to do? You guys in a pickup truck and 100,000 GPUs, what could go wrong? Let's put a pin and switch real quick, which is, what did we see? We saw something that if you built into that 1,600 acre land bank and you built into that power bank, we can basically light up at 7 million per megawatt. It costs everyone today 11 to 12 million, as you guys know, per megawatt to build. Because we have all the infrastructure at these massive campuses, there was just a massive pre-cash flow conversion story that nobody saw. Everyone missed it. Because everybody was like, my god, Rob needs so much capital, Rob needs so much capital, and something's wrong with Switch. There was nothing wrong with Switch. It just needed a ton of capital. And generally speaking, Rob didn't really enjoy being a public CEO. So I think that we got him private. We liberated him. We gave him a bunch of capital. We gave him customer relationships. We taught him how to securitize and boom, the business is Forex in two years. Like that's the cookbook. Hiten Samtani (19:14) Marc, this is what I warned you against. You're a very good salesman and I said we're not going to do that in this thing. Marc Ganzi (19:20) Yeah. Like I'm not interested in selling, but I think the differentiation is you asked me, what's the difference between me and Blue Owl or Blackstone or Brookfield? We can all raise money. We all have the same LPs, actually. I'll go to the Middle East and I'll bump into, pick the company. We literally bump into each other in the elevator banks, right? But we compete for capital. So how do I compete against Blackstone? How do I compete against Brookfield? And what I say is it's really simple. I've been doing this for 32 years. The other guys, are new to the street. And so you asked me what could go wrong with these real estate guys is they're late. They're just late. Hiten Samtani (19:59) So this, well, I think is Gansi's big edge as he sees it. People might have bigger checkbooks, but like he's the guy, he's been in the space longer. Will Krasne (20:06) If you have to stick around and you have equity that you're rolling, then yeah, you want to be with somebody who actually knows how to grow the business and can bring stuff beyond just capital because capital is a commodity. As Mike Comperato said, our great episode was black Hondas. And if you can come in and say, we're going to do these specific things and I've done them before here and listening to Marc's pitch, it's very convincing. Hiten Samtani (20:19) Black Honda Then I think what Marc brings here as well is kind of this is a nascent space, a lot of the capital structures are being formed as we speak and Marc is kind of a pioneer of structuring a lot of these things. And we close the chat by getting into one he did for CoreWeave, one of the largest ever private dead runs. You talk about freeing, let's switch gears and talk about the opposite, the lockbox, the core-weave financing. This is a fascinating structure. When I read about this for the first time, I know you're in partnership with Blackstone and a couple other players in this deal, put in unromantically, you're holding these chips hostage until they pay back the money. And the interest rates in this kind of financing are quite high compared to traditional financing. It's not quite pick financing, it's its own beast inside the machine. How the hell does a structure like that come together? Marc Ganzi (21:21) Look, if I could go back in time and rename it, I wouldn't call it like GPU revenue notes. I'd call it like AI factory revenue notes. And I probably would have priced it tighter. Like if I just like change the label on the box, you know, pull a page out of Tommy Boyd. But it's not on the box. should always be on Like we probably would have gotten 20 bibs tighter. But no, in all seriousness, we've been absolutely fascinated with the Core Weave story. We got involved with the company five years ago. We were standing up our credit business and we wanted to make Hiten Samtani (21:37) on the box. Marc Ganzi (21:51) a big name for ourselves in the credit world. And so we were talking to Magnetar, we were talking to Code2, and they had noticed, like others, that we'd done a ton of securitizations. And we'd figured out, I created the cell tower securitization structure in 2004. I pioneered the data center one, the fiber one, the small cell one. We basically have gone to the rating agencies many times, and we've been the pioneer of these kind of off-the-shelf esoteric structures where you're basically financing digital infrastructure. So with CoreWeave, it was interesting because usually I can attach to a data center or I can attach to a tower. can attach, I can put a. Hiten Samtani (22:31) There's real collateral. Marc Ganzi (22:33) the collateral package here was weird. so we said, okay, how can I always like when I do a securitization or private securitization, I say, how can I get the belt in the suspenders? So what do we have with CoreWeave? And so we're like, well, look, I think people can get excited about a seven-year contract with Microsoft. I think that's investment grade cash flows. The ARD data is five years. There's that famous slide in every securitization roadshow deck, which is like where the cash flows stop and where the ARD data is, right? And so the ARD day was here and the cash flows were here. So we knew we were safe. We had a year and a half to two years of coverage on the contract with Microsoft. So essentially we sort of pitched it almost as a Microsoft look through type of relationship. So you get the revenue from Microsoft as a pledge, you're pledging the revenues and then that's the belt and the suspenders is you get a lean on the GPUs. So if for some reason this thing goes horribly, horribly wrong, we step in and we take over the GPUs. Will Krasne (23:31) If you're taking over the GPUs, you know, can't come in, renovate, throw on a new roof, and go release. Marc Ganzi (23:36) ⁓ but you have a lender that owns like 380 data centers and knows how to run a data center. So stepping into their prime lease in the data hall, I can do that all day long. So for me, taking the keys back in that situation, I'm not that nervous. Why? I wake up every day, I fall out of bed and I run 300 plus data centers around the world. We don't have any fear about running the asset. First of all, we don't think we're ever going to run the asset. Hiten Samtani (24:02) there like an obsolescence risk to the GPUs? The problem with your business to us as outsiders laymen is the fundamental clash with Moore's law here. That's the thing that both of us are really grappling with. Will Krasne (24:16) How do think about reversion? ⁓ Because we just don't know. I mean, I don't know. You have a better view than I do, obviously. Like, what do think the world looks like in five years once all this AI infrastructure is built out? Marc Ganzi (24:28) This is the same narrative I got 20 years ago when I got into cell towers. There was like, ⁓ know, satellites are going to replace towers. And then when 3G is done, you got nothing and there's no residual on the towers. And I was like, okay, like that's a narrative. The reality is at every technology upgrade, at every iteration, I win because I don't really care in my data center whether you're using an H100 or a Blackwell chip. I don't. You got a 15 year lease to me. You're Microsoft, you're Amazon, you're Google, you're Meta, whoever you are. My data center is connected to like 50 to 200 other data centers that you own. It's a network. So one thing I learned a long time ago is that in digital infrastructure, 98 % of the time the customer renews. 98 % of the time. It's stunning. You'll never find an industry more reliable on churn than towers and data centers. Hiten Samtani (25:22) The cost are too painful or what? Marc Ganzi (25:24) Switching costs are massive. To pick up 100 megawatt data center in Santa Clara and move it to Reno would cost you $400 $500 million. Will Krasne (25:32) At what point though is an upgrade worth it? I'm in industrial for the most part in multi-family, so it's obviously way different business, but on industrial, it's like if I don't need to be in X location and I can go get a new fully automated warehouse in Lebanon, Pennsylvania versus being in South Jersey, that's X amount better and it's worth it to me. Is there a threshold to where it is worth it to spend that money? Marc Ganzi (25:57) For you, it depends on how tethered Lebanon is to the New Jersey warehouse. Are they tethered? Are they connected? Do they rely on each other? So a data center in Reno that's tethered to a data center in Salt Lake, an edge facility in Salt Lake, you can't sever that connection. They rely on each other. They're talking to each other. They're passing data workloads back and forth. Networks are living, breathing organisms, connected by fiber, connected by servers, connected by cell towers. And so these networks are super unique. which creates that stickiness in the renewals. Hiten Samtani (26:29) Even since you did the core we have financing, has there been more innovation in that space from the, from the debt side? We talk a lot on this podcast about the financialization of everything. this, like, when I read this piece, when I read this piece in the Wall Street Journal, I'm like, wow, this seems like the final boss of securitizing certain. Tell me what you're looking at now. I know you can't disclose any, you know, upcoming deals, but like, what are the types of things you're seeing in financializing these instruments that are interesting, that are scary. Marc Ganzi (26:45) We're not done yet. I don't think they're scary. Everything has a bit in an ask, right? Risk reward. So if you look at, for example, we just did this big Rezzi fiber deal with Allo, 550 million dollar financing. It wasn't a securitization, but it essentially is the bridge to the securitization. We're helping Allo get ready to get securitized and get out. And so what we did is we did a second lean across their network on all of their new fiber builds, all the homes that they're passing. so financing residential fiber is hard because churn's a lot higher. Churn there is like 10%. So the structure has to be right and the coupon has to be right. So that was an expensive loan, but it gave Aloe 550 million dollars of capital to essentially double their business plan. We're always thinking about how to test the boundaries, whether it's a private securitization, a bridge to a securitization, a second lien structure. Hiten Samtani (27:51) I think one of the things that you and I initially connected on was I saw someone who is not only pretty deep into the space that he's in, but also is probably one of the best people at kind of forming the narrative around that space. And one of the things I think about with DigitalBridge in your career is like, to be priced correctly, there has to be a strong definition, a strong narrative around your company. When you do all these things, what's the overarching umbrella for them? If someone wants to buy you out and then we can't talk about rumors right now. but if someone wants to buy you out, what are they buying? Marc Ganzi (28:27) What's interesting about us today is when we did our merger with Colony, we inherited a big balance sheet. We had like a 19 billion dollar balance sheet. Hiten Samtani (28:34) This was a major turnaround job, which is why you and I first connected. Marc Ganzi (28:37) Yeah, we had to kind of like reassemble it from the ground up and we had a bunch of real estate that was a melting ice cube. We had to sell industrial, we had to sell hotels, we had to sell medical, there's a bunch of debt funds we had to get rid of. And so we quickly dismantled $53 billion of assets and we got them sold off to different people really fast. We had a digital book about 14 billion back in 2019. And so we quickly got the business delevered, we right-sized the cost structure. And we just went all in on digital, but the only way to get there was to go asset light. We couldn't do it through the REIT because the size of checks we were writing were so massive. So we did it through funds. And we told people like, look, if you want to invest in the largest owner of digital infrastructure in the world, you got to invest with us. We're by an order of magnitude. We're twice as big as an American tower, a crown or digital reality and equinex, the amount of assets we manage. The only difference is what you're buying is a fee stream, which is our management company. So think of us as the digital gangsters. If Blackstone just did digital, that would be us. Or Brookfield just did digital, that would be us. Will Krasne (29:45) So what I think is really interesting though is that's 180 degrees from where you and Ben were when you left GDP. it's the exact, like you talk about asset light, there's nothing more asset heavy than, you know, dipping into your own checkbook and starting off and putting those things together. But is this where the ultimate end game of where you guys saw like going? Cause you could have just kept owning those assets cause you were so early to your point. And I would imagine there's like quite a lot of embedded value in there that you guys could have spun up a business just. through that and consistently recapping it and growing it that way. Marc Ganzi (30:17) We'd go too slow. I mean, the market was just moving so fast. And so the only way we could keep up was by raising third party capital. know, Ben and I did put a bunch of our net worth into the business the first couple of years, and we stood up a bunch of businesses on our own. But what we found out is that what everybody finds out is these businesses are really CapEx intensive. And so we had to get out there, we had to raise third party capital, and we did. And COVID came and everybody wanted exposure to digital and... We're just five or six years ahead of everybody else. Will Krasne (30:49) Do ever think of a way around getting permanent capital to where you don't have to be on the fund life cycle? Marc Ganzi (30:55) Yeah, that's like the Shangri-La. That's where you want to be. We are out launching a real estate strategy that will be a permanent capital vehicle to own investment grade, best data centers. It'll be our first strategy where we have permanent capital. Everyone copies everybody, right? Everyone's copying digital? Well, we're reaching in our playbook in private wealth and permanent capital and looking at guys like Apollo and Aries and how they form capital and we're just kind of following them and they're following us. It's sort of funny. Hiten Samtani (31:23) I'm going to close with ⁓ maybe a more personal question, non-technical. ⁓ To do what you need to do right now, given that you're at the center of one of the most rapidly growing asset classes, just requires a ridiculous amount of energy and keeping the faith at times when your business, long game business doesn't really work with a quarterly schedule, right? Like you're making these decade long 20 year bets. Marc Ganzi (31:50) It is funny actually, you're right. ⁓ Hiten Samtani (31:51) You got to go talk to the market every quarter and some quarters are good and some are not so good, right? we just want to understand on how you keep the faith, how you keep the energy and how you can kind of stay in cell mode for so long. Marc Ganzi (32:04) I lead a pretty simple life. don't spend a lot of time wasting my energy on things that don't really matter to me. I'm really brutally efficient with my time and I put a lot of energy into wellness. Hiten Samtani (32:16) ⁓ Does Polo count as wellness or? Marc Ganzi (32:18) totally. I mean, it keeps you sharp. It's exhilarating. It's fun. It's dangerous. Will Krasne (32:23) you know there's there's nothing better for the inside of a man than outside of a horse. Marc Ganzi (32:27) I didn't know that. Hiten Samtani (32:30) Marc Ganze, thank you so much for being with the Promote Podcast. Marc Ganzi (32:33) Good to see you guys. Hiten Samtani (32:36) Marc from Digital Bridge, super knowledgeable, super passionate about the space. That's it for the Promote Podcast this week. We'll be back next week with more CRE insider goodness. Thanks again to our sponsor, MADDPROJECT. If you're interested in their NYC resi development fund, reach out to antonia@maddproject.com That's Antonia at maddproject.com. Will Krasne (32:56) Like, Share, Subscribe. Hiten Samtani (32:58) Keep voting us up. The top 100 is already a distant memory. We want to be the top 50 next, so keep going. Will Krasne (33:03) like the best capital allocators. We're coming for you. Hiten Samtani (33:05) We're coming for you. To our listeners, thank you and well, thank you. Ciao. Dude, I feel like I want to go do something analog. Will Krasne (33:19) Yeah, that's why I'm only writing things on pen and paper right now.