Sure. All right. Wrapping up our morning fireside sessions, we are joined by Chris Ferraro, who's President and CIO of Galaxy Digital. Chris oversees principal investments, research, and corporate M&A, in addition to a central role in the firm's strategy. Since last May, Galaxy has completed its Nasdaq up-listing, delivered its first data halls at the Helios campus in West Texas, which is what management has called the single most important de-risking event in the data center business. You began repositioning the firm into a B2B digital infrastructure platform. Excited to get underway with you, Chris. Thanks so much for joining us.
Happy to be here.
All right. First quarter looked challenging on the surface. Crypto prices were down 20%, 25%. Balance sheet had mark-to-market losses. Lending book contracted, and yet the operating business held up. Trading volumes were flat against the industry, down more than 25%. Asset management was strong. You said on the call EBITDA was tracking pretty healthily in the second quarter. How would you characterize broadly what's happened in the digital asset space in the last 12 months? Does the resilience of Galaxy's model, as we saw in the first quarter, does that give you confidence that the business itself is becoming less structurally cyclical?
Yeah. We can go a lot of directions with this question. How I would characterize the digital asset markets, I think it's really important that we differentiate between crypto assets and the digitization of traditional assets. They get conflated a lot. Crypto assets, so when I say crypto assets, I mean like Bitcoin, Ethereum, Solana, as true, like decentralized crypto Layer 1, Layer 2 assets. Those have been challenged. They've been challenged for a lot of reasons. I think Occam's Razor would tell you that they have historically been early-stage venture bets on a new decentralized financial system that have been very momentum driven, very speculation driven. There's a lot of stuff going on in the world that people are excited to speculate on and take some of those momentum and speculated dollars out of crypto and go into AI and other big momentum trades.
I think you've seen capital flow shifts, and excitement shifts away from crypto assets into those other areas, namely AI. The thing that Galaxy has been focused on, and we benefit from, is we don't just trade and own crypto assets. We had to build digital infrastructure and services on top of digital infrastructure in order to do that. What's happening at the same time is many global financial institutions have decided finally that digital infrastructure is coming for traditional financial rails and feel very short. The technology, the infrastructure, the ability to transact, to hold, transfer, finance anything non-analog. The other side of the coin that is happening very fast is a very large emergence of TAM for us to help build for traditional institutions to operate digital infrastructure for digital assets, whether they be cryptocurrency assets or traditional digital securities now.
You see that in our results, our business lines have morphed over time to providing infrastructure that is a lot more stable. The opportunity set in the forward is going to be a mix of both the cryptocurrency assets as well as digital asset convergence. We're not immune from digital asset prices. It sucks when digital asset prices are down, for sure, for Galaxy and Mark, [Crescent], but we see the emergence of a very large TAM outside of that coming and we're meeting it.
Yeah. It's definitely a large TAM. On the call, you talked about how what was once your biggest competitors being the major banks are now potential clients and prospective customers. What's actually driving that demand, and could you maybe just articulate what you expect Galaxy's role to be with these major banks? What's going to be your niche in this tokenized world?
One of the big questions we were faced with years ago launching Galaxy Digital has been like, "Okay, this is great. For now, you guys have edge. What happens when Piper Sandler or Goldman Sachs shows up and wants to do what you do, and they've got 10x the workforce and everything." There's never really been a good answer for that other than we're going to keep iterating on the edge. Be better technologically, we'll be able to compete. That whole dynamic has flipped. One of the assets we bought back in 2022 that we've developed over time technologically, internally, have deployed, has been our Base wallet technology. At the core foundation of holding anything digital bearer asset needs to be holding digital key material safely.
That's an example of what every major financial institution now is saying, "Man, I need to figure out how to do that, like with zero risk of loss, or else I can't participate in the digital bearer asset version of the world in the future." That's an example of the type of thing that we own, we have capabilities in, that various institutions are figuring out whether they also need to own it, whether they can lease it, whether they can rent it, whether we can be a service provider. What I think is going to happen, and we don't know because market structure is developing daily now with things that are going on regulatory-wise, is I think what the role Galaxy is going to play for big institutions is going to be for the biggest institutions who want to and need to own their own technology.
I think Galaxy can play a big role in helping deploy teams to help the institutions build that technology, perpetually license under an ownership structure, technology that we already own, out to those institutions. Help operate those systems with those companies for years to come, and also plug our services on top of those on the other end. Think about on the right end, you've got the biggest institutions who would pay Galaxy to help them build their infrastructure and then be partners on the go forward. As you move down market to middle market and lower market institutions, Galaxy owning that technology and the capability can be a direct service provider. It's a question of build versus buy versus rent for those clients, and we're going to play up and down that stack.
When we think about the other aspects of the platform right now, you have custody staking, liquidity, fund products, lending. How big of a revenue driver in the grand scheme of the digital asset business could those more recurring licensing revenues or service provider revenues be?
Our goal is to make that the majority of our revenues on the go forwards. It's no secret to us that we think investors in equity capital markets generally like visible, predictable long-term streams of revenue as opposed to volatile streams, or even if you might make more money doing it in a volatile fashion. The goal for us at management has been to challenge ourselves to turn a episodic volatile market into a business that pays us an annuity stream. The majority is the goal target.
Sure. Maybe shifting gears, asset management is one of those businesses that can be pretty volatile. You just launched a hedge fund there. In asset management specifically, how do you expect going forward to compete against firms with dramatically larger distribution, and where do you see the differentiated opportunities there?
Yeah. I think we're not naive in that we don't intend to compete with the large asset management platforms on the planet who have the distribution. I also think that asset management is probably going through a pretty interesting time now where what I would call actually asset aggregation and collection probably will turn out to, in hindsight, to have been not the best alpha generation for investors. As a firm, we lean into where we have been good all along, is we are leaning into niche alpha products for ourselves that we launch directly for clients where we know we have edge, where we size funds at the appropriate size, where we know we can create great returns for investors, and we're going to build over a long period of time organically that way.
For what I would call more like access products that require distribution that can scale really large, our strategy has been what it's been historically, which is to partner with big institutions who have distribution, where we can add a lot of value on the underlying assets and creating the asset structures, and let that partner be the front face, let them own the product, let them own distribution.
Sure. Let's talk regulation. Galaxy's been one of those firms who's been very vocal in voicing its opinion and active in D.C. and being part of the overall crypto conversation there and digital asset conversation. On CLARITY Act specifically, it seems like it's 50/50 whether it's going to pass or not. How transformative is that legislation for Galaxy? What's your honest read in your seat in terms of whether or not Congress is going to get that across the finish line?
Yeah. I think we think it's better than 50/50. That number does move around daily, though, as new information comes out. I think it's wildly important for the digital asset industry for it to happen. I think what's happening, there's a few different camps. The big battle that's happening at the highest level is a reticence by traditional financial institutions who have regulatory moats and somewhat oligopolistic market structure, afraid of allowing innovation and products on the fringe to come in because they've seen how quickly things like stablecoins can go from zero to $300 billion of assets overnight. Those are the types of flow. Having the ability for capital to flow that fast into new products is a scary thing for an institution that thought they had their arms around being able to hold onto all those assets. I think cooler heads will prevail on that front.
I think ultimately, the government knows that this is good for the country. It's good for America to be more forward-looking instead of protectorate in terms of your old market structure. We think the probability is pretty high. I think it's wildly important for the market to happen. One of the biggest things that's held the digital asset markets back has been the inability for companies and digital assets to innovate and try to create new rails and new products because of the regulatory hangover. For us at Galaxy, we've never been a fringe player. We've never wanted to operate outside of the regulatory framework. It's held us back as an institution in terms of profit opportunities. It's also kept our reputation largely pristine.
Adding regulatory clarity, creating rails for us to actually operate is only a good thing from our perspective.
All right. We covered the digital asset business. I want to switch over to the data center piece of Galaxy's business.
Yeah.
Talk about Helios. Phase I right now is live. You delivered the first halls to CoreWeave. Phase II deliveries are expected the first half of 2027, and then you have an additional 830 MW that was approved by ERCOT recently, and you're working toward, I think, signing a tenant. Could you just give us the state of the play on the moving pieces at Helios? Also part of that that I missed was the additional 1.7 GW that I think ERCOT is still under study with ERCOT. Those three pieces, where are we at there?
As it relates to our asset at Helios in West Texas, I agree, those three stages are correct. The first stage, the first 800 MW, fully approved by ERCOT, energized. We've been flowing power through those lines since 2022 when we bought the asset. Bitcoin mining, we are developing that 800 MW for CoreWeave and a full build to suit build for them. You're correct. We are the first 200 MW of gross power will be energized and delivered for them relatively soon, by the end of this quarter. The remaining 400 MW and then 200 MW, the last 600 MW will be a rolling delivery in 2027 into 2028 for them.
That, for us, will create a stabilized asset with a +15-year offtake tenant paying us $1 billion+ a year in rent, which allows us to use that at near triple net lease type margins, which allows us to then have a really stable platform to take that capital and reinvest into growing the platform. That's one. Two, the next 830 MW, you're right. We got approval from ERCOT for that 830 MW. At the risk of getting a little too nuanced, think about that 830 MW as having now been fully approved at 100 MW allocated to Galaxy before the new batch process with ERCOT, which is now going to piece out pro rata allocations to people on the go forward.
We got a full allocation of 830 MW that we were applying for right under the deadline before ERCOT has now switched to a, let me collect all the market participants, evaluate what their needs are, and then parse out what's available to them on a pro rata basis. That's fully allocated to us. We will make the attestations to ERCOT in the next month that they newly just came out two days ago with, are going to require for that. That has given us the confidence to say, "Okay, that's ours." We can now go out and have conversations to lease that capacity for a build. That's deliverable to end clients in the end of 2028, 2029, and 2030 is when we can ramp up power at that new station.
The final 1.8 GW minimum, I think it actually might be higher now, that we have under various forms of study are going to be a part of ERCOT's new batch process. We do expect a portion of that is going to be a part of batch zero. That's still under debate with ERCOT. No one has full clarity on that yet. The idea is that over the next decade, that near 2 GW of additional power, in addition to the 1.6 GW we already have granted to us, is going to be approved. We're going to fund CapEx into the grid to help build stabilization so that the grid can accept that. Over time, we're going to grow the Helios campus to a 3.5 GW , 3.6 GW mega campus for hyperscaler tenants to come in and run their GPU clusters.
Any update you can give us on how conversations have gone with potential tenants for that 830 MW?
Yeah. Demand is very high. The one thing I would say, tempered a little bit, is we've been very clear. The new 830 MW that we have, which is not allocated to CoreWeave yet, or not allocated to CoreWeave as like the first 800 MW is 2028, 2029, and 2030 power. The real conversations with the buyers of power are they are actually very focused still on trying to fill their holes for 2026 and 2027. 2028 for them, they know they're going to be short power based on their expectations in 2028 and 2029 and 2030. They actually have to solve 2026 and 2027 problems today. Our goal has been make sure we develop the right relationships with the biggest companies in the world so they know who Galaxy is. That's actually a pretty big lift.
Make sure we're on their radar, make sure they know what we have, and develop the relationship so that when they decide they really need to now move into 2028 power, we're ready to go with that conversation.
I don't want to put you on the spot, but I will.
Yeah. Go for it.
When do you think, from a timing perspective? 2028, 2029, 2030, when are they looking to get that power? When do you think that time will be that they start having that conversation?
Yeah. I think the best way to characterize it is an interesting cat and mouse game between them, ourselves, their internal teams, right? It's no secret, they all have internal land development teams, right? Where Microsoft and Google and Meta have lower cost of capital than Galaxy. That's not up for debate. There's this dynamic where it's like, well, maybe our teams can locate an asset that you guys already own, and we can develop it ourself in time, and that might be cheaper overall, and we want to hold up the optionality for that. That is the dynamic, not just with Galaxy, with every sort of non-end user who has access and owns power and is proposing to provide data center services to these companies. That's what we're dealing with.
I think history has told us that their expectations around their own ability to self-develop have been wildly overestimated over and over again. We know that we own the asset, which is great, and so we just want to make sure that we build the right relationships so we can make the right decision to pick the right partner when the time is right at the right economics.
Yep. All right. You've built a meaningful infrastructure at Helios, but you've also discussed going beyond the Helios campus and looking at other sites. Mike mentioned it on the call. As we think about the next leg of growth in the data center business, what does the pipeline look like for land development and land purchases away from Helios and also power acquisition opportunities. I'd put in that as well.
I think we benefit from the fact that we can use Helios as a base, a base load asset, right? We created Helios from an original purchase price of $65 million back in 2022. That base allows us, we have forward expectations of available capacity up to +3.5 GW that we can develop off of that low-cost basis that we own, and we own the outcome of, right? We do have ambitions to have a diversified portfolio geographically, tenant-wise. Expanding horizontally away from Helios makes a lot of sense and something we want to do. The market today has been, we've spent a lot of time with a lot of potential opportunities.
It's filled with a lot of land speculators who have tried to take control of land, don't have the capital to develop it, don't have the expertise to develop it, but are playing off of the theoretical insatiable demand and want to flip those assets for very large numbers. We're just not in the business of spending hundreds of million dollars for spec-powered land because we already own a big base of asset that we can use ourself for that kind of development that doesn't cost us very much. We've been very selective about things we're going to actually move forward with. We're kind of letting the market shake out who's real and who's not because, and ERCOT's a perfect example.
Like in Texas, ERCOT has basically said, "We know this is happening, and now if you want to apply for land, you're going to have to put up real dollars." Every market participant now on the forward and the back process is going to have to put up $50,000 per megawatt of security just to apply for power. Those are kind of things that have turned into opportunities for us where because we have that base asset that we can leverage off of, we're now a go-to participant in that market to partner with to actually take over a site to develop it. We would've hoped that there were more rational economic actors that we can grow the land portfolio horizontally faster. We're just not willing to do that at stupid prices on spec. That's it, yeah.
Helios was definitely not a stupid price. It's one of my favorite stories in the space. I think you bought it in, what was it? 2021 for like $60 million?
$65 million .
$65 million in an opportunity zone, and now it's this enormous multi-billion dollar data center site. It's quite a trade. A question that I get asked quite often is, why don't you just separate these two businesses, the digital asset business, the data center business? They're not naturally synergistic. I think on the last earnings call, Mike said that any separation decision is more of a year-end debate than a right now debate, if I'm not mistaken. Two very different investor bases for these business, different valuation frameworks. How are you thinking about it internally, and what changes between now and year-end might inform your decision of whether something like that makes sense?
I think from a trading perspective, I'm not a trader, but Mike is. From a trader perspective, there's a lot of instinct in a down crypto market to be like, "Oh, it's obvious you should separate those businesses," right? I think if we were in a up crypto market regime, that noise would be a lot less, right? From our seat managing the company, we say to ourself, these aren't trades you can just unwind. They're not easily separable assets. We have mind and management that sit over both businesses. There's expertise in not necessarily in the construction of data centers, but in the structuring of lease deals, in the financing that is part and parcel and at the heart of what our core management team does in the markets business.
Historically, it's made all the sense in the world for us to keep them together. I think there's good, accurate criticism of, well, if you're not getting full credit in your price, then your cost of capital is too high. We've already raised the capital we need to build out the first section of Helios, we don't really have a capital raising need. I also think that we've wanted to have the optionality, like there are adjacent parts of our business. For example, if we're a multi-hundred million dollar natural buyer of power every year, and we have a global trading business, and we haven't traded power yet, now you're industry, you have an edge in trading power like every E&P company has done historically.
There's early stage compute markets that are now starting to emerge in terms of compute derivative markets that we think are actually kind of interesting and fit really nicely into our trading desk to potentially warehouse and trade compute derivatives and compute risk on GPUs. There are things emerging that I think if you just agree to separate the businesses, you really give up optionality on that we're just not excited about giving up that optionality yet when we don't have to. It might become a natural thing for us to do. It also, like I said, things might evolve in a way that it doesn't make sense. That's how we're looking at it.
Yeah. That is interesting. There's a synergy between digital assets and the trading business and the data center business compute perpetual futures. Compute derivatives. There you go.
In the last three months, they've already emerged and there's started to be--
ICE, CME
There's started to be 10-figure trades that are happening, and it's very new, and whenever that happens, there's huge early participant opportunity.
Sure.
We're supposed to meet it.
Yep. All right, Chris, that's all the time we have, but thanks very much for joining us.
Thank you.
All right, everyone.
Thank you very much.