Fred Thiel: Sure. If we had historically been a kind of de novo site developer, I could give you a whole pipeline with a chart of okay, these sites will energize on this bump, et cetera. That’s not been our historical model. Our historical model has been as, you know, agile, grow quickly. And we’re now in the business of being more of an owner-operator and vertically integrating. So you can think of sites across three buckets. There is the bucket of, I’m going to go buy things like Granbury and Kearney, where there is excess capacity with an ability to grow. And as the hosting customers at those sites age out, as those contracts age out, we will absorb all that capacity ourselves. So we’re looking — think of it this way. You have a site with a, I mean, 100 megawatts, maybe 80 megawatts is used, okay, we can plug 20 megawatts of miners.
And then over the next two years, half of those contracts for the 80 megawatts will age out, and we’ll start deploying more and more miners there. And oh, by the way, that site could potentially add 100 or 200 megawatts more because of the substation and we’ll develop that and add that capacity. So not to give everybody our playbook, you can send me an email. Happy to send it to you. But that’s how you look at that bucket. If you look at the next bucket, it’s sites that somebody may have permitted, somebody has gotten some form of allocation of power. There is a substation available. They may even have transformers underground. But they haven’t built site yet because they don’t have money. And this is kind of back to Reggie’s question. There are people who have sites blocked off.
They can’t raise the money to develop them because investors aren’t willing to give you money in today’s kind of market for Bitcoin mining datacenters because they prefer to give money to people building AI datacenters because there’s a lot more money to be made there, if you believe people who believe that that’s the case. We happen to believe Bitcoin mining is the place to be. But we’ll let the rest of it would be up to contention. So those are kind of — think of them as halfway down sized, somebody — it’s kind of like in the real-estate development world, it’s not raw land. You actually bought the land, got it entitled. You’ve laid the sewage and utility lines. And now you’re going out calling to the homebuilders and saying, hey, you know, I’ve got 16 lots here, you want to build a home?
So that’s kind of middle bucket. These are sites that could be energized and online in a six to 18-month window. Then you have the true greenfield sites, where we have folks that are out today scouring opportunities to acquire access to power, access to land, access to transmission interconnect, et cetera. On a global basis, not just in the US, and those are kind of 18 to 24, 36 months type projects. And so we believe that as a global world-class Bitcoin miner, you need to build a stack, if you would, of staggered projects that give you immediate capacity, mid-term capacity and long-term capacity but that all have optionality. So I’ll give you an example. Granbury, Texas, has about almost 300 megawatts of capacity today. There is an opportunity to expand it because the power station has lots of power.
So that gives us optionality. We could add more capacity to it if we want to invest the money to it. So you already have a site, you already have a power partner, you already have access to substations, et cetera, it’s just a question of when do you want to start developing it and how much money you want to spend to develop it, and what are you going to use that site for, immersion, air-cooled, whatever. With the longer-term sites, the optionality is it’s very inexpensive to tie-up an option on long-term power, access to substations, relatively speaking, when you talk on a per megawatt-hour basis. And lease land. And you can sit there. And this is the business that, in the old days, the Compute Norths of the world used to do, which is they would go out and they would tie up a deal with the power company and then go find a miner who would essentially be willing to fund the build out of the site, and then 12, 18 months later you plug in.
The King Mountain site in Texas was done that way. All right, reengaged with Compute North back in day, they engaged with NextEra Energy, and they got the power, they got being permitted, da, da, da, da, da, and then they built the, site and then we came in. In this case, we’re acting as the builder operator, right, of those sites. And so we’re working along all three of those tranches, if you would today. With partners and directly ourselves. And so we — our goal is essentially to have a think of it as a store house full of either, readily available immediately today mining capacity, mid-term available capacity. There is an option for us. We know that when we needed 12 months out, we just turn the crank and it will be operational, already permitted, you already have all the transmission, et cetera.
And then the longer-term sites where we want to build 100 megawatts, 200 megawatts, 500 megawatts at a location where we may have 700 megawatts of potential capacity, if we’re willing to do the longer-term investment. And especially internationally, those longer-term — that last bucket, there is a lot of available opportunity there, Ethiopia, Paraguay, et cetera, et cetera. So, the goal here to build a really resilient business. As you build a huge pipeline and lock up potential capacity sites, with optionality to it. And by the way, I’ll mention one thing, the Granbury site, for example, the way the PPA works there is it’s — it’s not like we have to take all the power. We can actually, in the event of really bad pricing in the marketplace scale it back.
So we have ultimate optionality there, which is the best thing, right? So you look for creating a portfolio of capacity that is short, medium, and long-term. Excuse me. You look to have a portfolio and access to technology, which is short, medium and long term. And I’ll touch on that bucket right now. Short term is Bitmain ship me S21s. All right. Medium-term is MicroBT, Bitmain, Canaan, I want to access to your chips. I’m going to lock up a supply of chips, you’re going to sit on them for me, and then I’m going to tell you when I want you to build the miners. This is a very different model. This is a model from the PC industry and the technology industry. And no small player can do this. You have to be able to write a $50 million, $100 million check to one of these people who say, I want so many wafers of your chips.
And then when I tell you to, I want to turn into miners. And in that way, I can lock up capacity and I have no worry about getting access to take the chips. And the last one is what we’ve done with Auradine, where we actually own a part of the company who is designing the chips. Why is that important? Because we can get miners that have specked specifically suited to our needs and use cases. And when you see the two phase immersion technology that we’ll be releasing later this year, you will see the benefit of that where any other traditional miner, whether it’s Riot or CleanSpark, if they don’t know how to build technology products, they are going to be buying off-the-shelf PCs, when we’re busy building custom built high-performance systems.
And that’s the differentiator longer-term that we believe is the biggest moat with these guys.
Kevin Dede: I’m very much looking forward to seeing Auradine’s product in action. So thank you for that color. Can we peel the onion back just a little bit more though, Fred. Based on the numbers that you offered this afternoon, 22 exahash, you’re 24 now. Does that mean you’re at 27 by the end of March. Does it mean you’re — or if you’re going for 35, does it mean you’re at 40 by the end of June? How should we think about how that capacity actually comes online through the year.
Fred Thiel: So we’re already at 27. We were at 27 at the end of the year. Pretty much.
Kevin Dede: Okay.
Fred Thiel: So I think, the way you have to look at it, Kevin, I’m going to lay it up for you because we are going to play the game here where, in 2022, we said, we’re going to deploy this hash rate, we had these machines and that I was getting the question, hey, are they sitting — are they plugged in yet? Are they plugged in yet? Are they plugged in yet? So, we’re just going to talk about stuff when it goes live, going forward. So we’re giving you an idea as to what our pipeline of equipment is. Unfortunately, I can’t help you model the when because you’re going to see it will come in very interesting lumps, not smoothly. But when it comes, it’s going to come a combination of at a rush and then in blocks. So wish I could say more, but okay.
Operator: Thank you. Our next question is coming from Lucas Pipes from B. Riley Securities. Your line is now live.
Lucas Pipes: Thank you very much, operator. Good afternoon, evening, everyone. Fred, my question is around the capital budget for 2024. What is it and what should be able to break it down between miners and infrastructure? Thank you very much.
Fred Thiel: Let Salman answer that question.
Salman Khan: Yeah, Lucas, it’s — I think we talked about it last time as well. This is what we’re looking at from a — from a total capital perspective with the targeted growth that Fred talked about earlier today, somewhere around north of $200 million, so somewhere between $200 million to $245 million, somewhere in that range. And that includes — that includes our miner purchases, and roughly, approximately $180 million or so. Just a quick reminder, we have been buying and paying for some of these miners, so some of those payments may have already happened. I’m just talking about the accounting capital here. In terms of the rest of the stuff we have other, you know, technology businesses, other ancillary businesses that Fred talked about, a small portion of that will be allocated to that.
And then on top of that, we also purchased Generate’s assets in Kearney and Nebraska and also Granbury, Texas, and that will be added to the capital which has sunk cost at this stage, but that was about approximately $180 million.
Fred Thiel: And that, of course, is separate and on top of the other numbers you mentioned.
Salman Khan: Yeah, yeah.
Lucas Pipes: And I heard it right, it’s kind of $200 million to $245 million for 2024 and about $180 million of that is for miners.
Salman Khan: That is correct.
Lucas Pipes: All right. I appreciate it. Thank you very much. Good Luck.
Operator: Thank you. Your next question is coming from Brian Dobson from Chardan Capital Markets. Your line is now live.
Brian Dobson: Thanks. Thanks for taking my question. So you mentioned as we head into the halving, it’s very likely for smaller miners to be pushed out of the business. Do you have a view on the potential magnitude of the decline we could see in global hash. And could that decline potentially be offset by some of the other players you mentioned like nation states or large private entities.
Fred Thiel: Great question. So one way to look at this is look at the nonce analysis work that a number of people have done that’s been published and readily available, where you essentially can see the amount of hash rate coming from what category of machine, and in some cases, locations, and even energy prices approaching data. If you have the ability to do that analysis. So there is a — the industry average efficiency today is somewhere around 30 joules per terahash, 30, 33, kind of varies depending on how many machines are on and I think you’ve been time, when you look at the nonce analysis. So at 33 joules per terahash and Bitcoin at kind of 55,000, post halving you’ll likely see anywhere from 11% to 18% of hash rate come off.
Now it may not come up all at once. You know, some people may have a few million dollars in the bank and they’re willing to say you know what, I’m not going to shut off because I’m going to let the other [indiscernible] shut off so that I get the benefit of the hash rate dropping and now I’m profitable again. And so I’m willing to take a loss for a month or two or three or four maybe. So I would kind of say, if you have to look at kind of the window kind of halving, and then the first six months post halving. That’s when you’re going to see kind of the people who are hanging on for dear life are going to hang on as long as they can, other people are going to say, no, I’m out. I’m going to shut down and wait for better days, wait for Bitcoin price to go up or not.
Which you are not going to see based on the announcements of ourselves and our peers is the rest of the world slowdown. And so you may very well have a world where, by the end of this year, global hash rate is 20% or 30% higher than it is today. You may have a world where the exact same as it is today. What I will tell you is that, obviously, it’s very dependent on the price of Bitcoin. And the fact that — and I’ll go back to the thing about the ETF. With the ETFs sucking up Bitcoin — and realize nobody can print Bitcoin, people talk about supply shock at the halving, yeah, you go from 900 Bitcoin to 450 a day. So what? 450 Bitcoin date and going to do anything to the supply in the marketplace. What is doing it, what is creating a problem is the ETFs as they continue to vacuum up Bitcoin, and I don’t really see it abating.
It may not grow, but I don’t see that abating, based on the conversations I’m having with institutional investors certainly is that the available supply of Bitcoin in the market is going to start drying up. It already is. It’s at record lows on exchanges. What that does is causes huge volatility swings in price. All right? That’s where somebody at night when ETFs aren’t buying can short Bitcoin, drop the price down, because there’s no real demand. And then in the morning when demand comes up, the price comes back up, and there are already traders doing this. You may have seen a couple of days last week or the week before where Quant funds went in and 10 times the volume that a couple of ETFs we’re doing as a test for this exact strategy. I’m going to go short Bitcoin, spot Bitcoin outside of the ETF markets, drop the price, buy it up and let the ETFs come back, drive it right back up and do it again.
So you are going to see a market that is going to be having a lot of gyrations in it. And when that happens, some miners will say, oh my, God, Bitcoin price dropped, I’m going to shut down. Oh, my God, Bitcoin prices are going back up. I’m going to turn on and you’re going to see hash rate bouncing up and down, up and down, up and down. And it’s going to make for a crazy world. So I think some miners are going to become sporadic miners, if they have the right type of energy pricing contracts. There are miners that have PPAs, where it’s take or pay, they have to buy all 50 megawatts of the energy they’re contracted to, especially hosted miners. And let me take a minute and say that the third-party hosting business is dead. Nobody who’s hosting a S19j Pros and paying $0.065, $0.07, $0.08 can be in business anymore.
Unless Bitcoin just goes on a real terror because they just won’t be able to afford to do it because they’re not getting the economic benefit of curtailment. They can’t subsidize their mining costs through other means. And so it’s the third-party hosting business, the retail hosters, other than people who to host because they do it for vanity purposes. I really don’t see that business. And I get calls very frequently now from people who are hosting miners in location say, hey, do you want to buy my hosting contract, do you want to buy the miners I have plugged in here and just take them over for me? And so, I foresee that business dying. And so that will be a certain portion of hash rate that temporarily will come off. But by the same token there somebody like me saying you know what I’m calling every third-party hosting guys saying, hey, listen, do you want to — if you have an empty shelf, you know, I may be interested in buying the shelf from you.
I’m not going to pay your hosting fee. But all by the shelf from you. And I may buy you if you’re interested in it. And so this is where I think you’re going to see the consolidation. And this is where balance sheet makes — is so important. And just availability of cash. Anyway, I could go on for a long time on this.
Brian Dobson: Yeah, no. I think, it’s just a quick follow-up to that. So is that where you see, call it, the lion’s share of appealing M&A post halvings in that third-party hosting segment?
Fred Thiel: To some extent, again, who can afford to do it. I don’t see — Riot, CleanSpark, ourselves, yes. But Core most probably not. Core’s business model is predicated on a $0.07 hash price. We’ll be at $0.04 at halving. So I think miners with a lot of debt, miners have balance sheets that are kind of wonky can’t raise capital aren’t going to be able to do much. And in the case of Core, they have a big brother whose name is Bitmain who is able to plug miners in that you can do a kind of a rev share deal that way. But that’s not a way to service $700 million of debt. So, yes. The large scale hosters are, other than Riot and Core or kind of disappearing and converting to self-mining and the small-scale hoster, they can afford to buy their own miners. So I don’t know what they’re going to do. And if they don’t have good power prices, nobody’s going to want to buy them.
Operator: Thank you. We’ve reached end of our question-and-answer session. I’d like to turn the floor back over to Charlie for any further or closing comments.
Charlie Schumacher: Thanks, Kevin. Thank you all for your time today. If you have questions that were not answered during today’s call, please feel free to contact our Investor Relations team at ir@mara.com. Thank you. And enjoy the rest of the day.
Operator: Thank you. That does conclude today’s teleconference and webcast. You may disconnect your line at this time and have a wonderful day. We thank you for your participation today.
