Greetings, and welcome to MARA's Q4 2024 Earnings Call. At this time, all participants are in a listen-only mode. [Operator Instructions] Please note that this conference is being recorded. I will now turn the conference over to our host, Robert Samuels, Vice President, Investor Relations. Thank you. You may begin.

Thank you, operator. Good afternoon, and welcome to MARA's Fourth Quarter and Full Year 2024 Earnings Call. Thank you for joining us today. With me on today's call are our Chairman and Chief Executive Officer, Fred Thiel, and our Chief Financial Officer, Salman Khan. Certain statements made during this call may be considered forward-looking statements within the meaning of the federal securities laws. In particular, any statements about our future growth plans and performance, our liquidity position, our growth opportunities, and our future financial performance are forward-looking statements. These statements are identified by the use of words such as anticipate, believe, estimate, intend, design, may, plan, project, would, and similar expressions or variations. Investors are cautioned not to place undue reliance on these forward-looking statements. All forward-looking statements made on today's call involve risks and uncertainties. While we may elect to update these forward-looking statements at some point in the future, we have no current intention of doing so except to the extent required by applicable law. Our actual results and outcomes may differ materially from those included in these forward-looking statements as a result of various factors, including, but not limited to the factors discussed under the heading Risk Factors in our most recent annual report on Form 10-K and any other periodic reports that we may file with the Securities and Exchange Commission. Finally, please note that on today's call, we will refer to certain financial measures that were not prepared in accordance with generally accepted accounting principles in the United States, including adjusted EBITDA and non-GAAP total margin. MARA believes these non-GAAP financial measures are important indicators of its operating performance because they exclude certain items that we do not believe directly reflect our core operations and may not be indicative of our recurring operations. Please refer…

Perfect. Thank you for having me here today and excited to have a good dialog here. Maybe to start, I know you guys talked a bit about AI and expectations to give us more detail on what your plans are for next year. But a lot of your peers in the space are kind of looking to a co-location model where they're maybe spending CapEx to build the infrastructure and leasing them out to a tenant. Is that something you are interested in or are you just interested in helping, call it, balance the load at those locations?

I think a couple of things. One is, while there is a huge demand from AI hyperscalers or let's just say hyperscalers and AI HPC vendors because there are some that where AI isn't the primary load, for capacity today, a lot of those sites that they're looking for are 500 megawatts and more. I mean, there's been talking the press recently about the real focus from the hyperscalers on these larger sites and a number of our peers in the industry have even talked about this in their most recent earnings. We believe that as this build-out occurs, essentially space becomes a commodity, right, and when you're providing power ping and rack space essentially, you really are a commodity. And I think if you go back to the time in the first Internet boom, you had the huge data center build out and then you had an overbuild of capacity. And then as the next wave came, the second wave, companies like digital realty, dev-ops, and others who really became experts at this business. We think with AI, the trend is slightly different. There is a need for large learning centers, but the capacity in those centers really is shifting in that now people are very much more focused on reasoning models and AI agents. And this is going to shift the demand for what these sites do and how they're operated. And I think you'll continue to see in the press more and more people talking about how people are moving away from the need to scale already expensive training runs and moving more towards reasoning systems. So we don't want to be chasing a ball that we think is moving very quickly and where investments are going to have to shift. The power needs per rack are hugely increasing with each generation of Blackwell and it's going to be very hard for a lot of our peers, I think, to keep up. We believe the real profit in AI is going to come from inference. Inference is where you actually make money from AI. Once you have a model, you load it into an inference site. And that inference site typically is located in close proximity to a customer, not miles and miles away in the areas where you have low-cost energy. They need to be near where the customers are because the customers need low-latency, the customers need rapid response and the customers don't want the data to leave their control often. And so small inference AI sites, we believe are the future and we want to be able to provide a complete solution to customers for inference AI at the edge. And that's something we're going to be very focused on as opposed to just trying to scale and build very big sites where you're having to deal with customers with leases who are going to be putting in their own equipment.

Understood. Very helpful. I feel like 2024, as you guys noted was a transformational year going from a largely hosted Bitcoin miner to a vertically integrated miner. In your prepared remarks, you outlined that you expect to accelerate that trend in 2025, should we expect you to go out and acquire as much power in 2025 as you did in 2024? And any thoughts on regions or locations of where you would look to do that?

That's a great question. So we've stated previously that we expect to essentially have 50% of our capacity coming from international markets by 2028, and we think there's an important difference in the US domestic market and the international market as regards how energy infrastructure and the energy industry operates. The US is a highly fragmented market with lots of generators and distribution of energy, IPPs, all sorts of other providers. And it's a market where it's very fragmented. The international markets are centrally controlled. And if you look at what we did in UAE, which is by far one of our most successful projects when it comes to collaborating with an energy company, we went from contracts with the sovereign to a fully operational 240 megawatts of capacity across two sites in just under 13 months. That is what happens when you can partner with an energy company. And we have begun speaking with energy companies internationally about the opportunity to not be vendors of power to us, but rather to be partners in joint ventures. Where energy price is less of an issue, you can view that as a capital contribution. We're focused on building very large-scale opportunities internationally together with energy partners, where we can help them balance their grids, help provide them with curtailable load so that as they begin to add things like hyperscaler sites and AI, they have the ability to provide curtailment for those sites, allowing those sites to use extra capacity on the grid. You may have seen the Duke University study, which recently came out, which said that the US had 78 gigawatts of additional capacity if the load on it was curtailed 5% of the time. And so we believe that in centrally controlled energy markets, partnering with the energy provider and be in business with them, not being a customer of them, is the way those future markets will develop. In the US, however, we will continue to focus on acquiring energy assets because we believe being our own generator of energy and getting as near to zero cost as possible, which is critical to making Bitcoin mining viable for the long-term. We have another having opportunity coming in not very long time and another one four years after that, and grid energy is only going to get more expensive, and so miners will have to have extremely low cost energy and be in control of their own energy assets, or as I said back in 2021, it might disrupt, you'll have to be a partner with the energy company and not a customer. So you'll see us grow internationally using one model, you'll see us grow domestically using another, but over time, our sites will be split about 50-50 between the two markets.

In terms of owning the energy assets, how does that compare to maybe just, I guess being a vendor instead of owning -- in terms of scalability, like how quickly can you spin up exahash if you're going down the energy asset ownership route?

Here's the interesting thing, if you look at this wind farm in Texas that we acquired, for example, that's a wind farm that sells energy today, so we can scale into that site and still be generating cash flow from the site. And what we end up doing is optimizing every electron from that site. Wind and solar assets aren't able to sell 100% of their nameplate capacity 100% of the time. So only during certain peak periods where they really can maximize profits and these sites have to be able to be very profitable the few times that they're able to actually sell energy with high margin because as you're familiar with the duck curve, there's negative pricing due to all the intermittent generation that exists in places like Texas during certain times of the day. But if we instead can utilize every single electron that comes off that wind farm, we're able to maximize the profit per electron, and so it allows us to really be almost an energy trader where we're our own customer, but we can still remain as a vendor to the grid. So we don't have to build a site and fully populate it, we can do it over time, and the strategy of having very low-cost energy, which wind is, there's almost near zero marginal cost to generate wind energy. It allows you to, more importantly, take machines that have been running grid-connected and extend their lives by now moving them to this very low-cost energy site, which may not run full time, it may run four hours a day, it may run five hours a day. But because those machines are fully depreciated and your energy cost is significantly lower than grid attached, you're not able to continue to generate hash rate with a machine that's already depreciated at a much lower cost per exahash and extend the life, and we believe that lowers maintenance CapEx, which lowers your overall CapEx and makes you more CapEx efficient, which in the long-term benefits shareholders because you're not having to spend as much capital to maintain your capacity, which means that you don't have to do things like take on necessarily excessive debt or issue excess amount of equity to be able to continue to grow and stay competitive.

And Brett, I want to point out that historically, we have -- we used to be a 0% owned and operated company beginning of last year, and we exited the year at 70% owned and operated. If you look at our matrix, we are very focused on value creation for our stockholders, which means that return on capital employed is an important matrix for us, and our gross ROCE is more than 30% on an annualized basis versus our competition, which is single-digit. So when you think about acquisitions, we've used historically to -- we've acquired these megawatts and electrons and we've used our MARA playbook by improving the sites to our standards, improving the capacity. So, for example, at Granbury, Texas, we increased our capacity by 200% with hash rate going -- growing by 200% from Q1 to Q4. In addition, we also reduced our cost per petahash per day by 45% on this. So all in all, while our eyes, as Fred mentioned, is towards the future, when you look back, we are very careful of the acquisitions that we do. They are mindful of the use of the capital and also it actually unlocks more value for our stockholders ultimately.

Maybe just one more question on the power front before moving on. The 25 megawatt partnership with NGON, I guess, how is that progressing and how should we think about this model scaling past the initial 25 megawatts? At what point would you have enough information to make those decisions?

Yes, so the 25 megawatts will be fully operational by the end of Q1 this year and it's progressing as planned with no hiccups, and we think that it's going to be a great success for us. This has taught us that the ability to generate on-site is something very practical actually. And when you look at the challenge at the other end of the data center spectrum with the hyperscalers are looking at doing is they're all looking at generating their own electricity. And so there are opportunities for us to scale this along areas where you have access to natural gas in addition to flared gas. You have gas that's essentially stranded in pockets where you can deploy larger-scale generation and take advantage of generating your own energy and mining Bitcoin. At the same time, if you build those assets in places where you have AI optionality, they become even more attractive. And so we believe, in the long run, owning these types of energy assets are going to be a very valuable asset for the company as we continue to expand our footprint in Bitcoin mining and at the same time as we begin to enter the world of AI.

Perfect. Maybe switching gears to Bitcoin a bit. it's been past tough couple of days as there seems to be a bit of a risk off trade out there. I guess longer-term network usage and adoption and those fees are going to be more important, I guess, when you look at the Bitcoin ecosystem, what specific projects or use cases are you guys looking or tracking right now that kind of get you the most excited? And then as a follow-up to that, Bitcoin has certainly proven to be cyclical, what part of the cycle do you think we're in?

Well, I think Bitcoin has become highly correlated to equities. The growth -- the rapid growth of the ETFs is certainly proof of that, and if you look at the outflows from the ETFs over the past few days, it clearly shows that it operates very similarly to the equities markets. And so when Bitcoin is correlated to equities, the ebbs and flows in and out of Bitcoin are going to be the same as the stock markets. And you have recessionary signals hitting now, you have yield curve inversion again and you have a belief amongst consumers that inflation is running high still and that there's risk. If you talk to the average person on the street, things are still expensive out there. Prices haven't come down. And so I think you're starting to see more and more of a risk-off environment and that's an environment where Bitcoin typically sells off. You're also seeing the dollar still very high and Bitcoin and the dollar tend to move in opposite directions. And you're seeing gold really being a store of value that people are focused on. One of the reasons is you have governments buying so much of that. I think the catalysts now for Bitcoin are really going to be the -- you have 24 states in the US that have bills on the books to essentially create Bitcoin strategic reserves. If they do 2% of their assets in Bitcoin, that will have a significant impact on the market and the price of Bitcoin. And if the federal government follows suit, then that would also have an impact, especially as other countries will continue to do the same. So I don't think the long-term trend is broken at all. I think we're still very much in the early stages of institutional adoption of Bitcoin. What I think is exciting in the Bitcoin ecosystem today outside of Bitcoin as an investment instrument us and as an asset, all the applications that are being built around side chains where you're starting to look at monetizing or rather tokenizing real-world assets and you're starting to build use cases for Bitcoin that don't necessarily have to do with payments or investment assets. So I think the exciting stuff is yet to come when it comes to use cases, but there's a lot of stuff certainly being built. I think the world of Ethereum has certainly been suffering and it's become more and more of centralized -- its appearance is that more and more of a centralized network and under kind of central control. And I think that Bitcoin is viewed as the premier and most secure network that people want to build on and tools are being developed, and I think we'll see a lot of exciting stuff come even from the traditional TradFi over the next two years.

Yes, I agree. I think Bitcoin is where all the capital is and having the capital there makes it the most preferred destination to build on.

Yes.

As you look at your Bitcoin holdings, and I know over the last quarter, you guys made a meaningful evolution of your capital market strategy in issuing a couple of converts using those converts by Bitcoin. So maybe if you could just comment on that strategy and why you chose to issue the converts to buy Bitcoin maybe rather than go out and buy energy generation assets or additional miners or grow your infrastructure. And is that a part of a longer capital markets strategy that we should think of? Is this something you would expect to do on a more frequent basis?

We started buying Bitcoin to put on our balance sheet back in early 2021. And then we paused for a while because we were so busy and focused on growing our total exahash in our capacity. Last year, in the back half of the year, we had a theory that Bitcoin price was going to appreciate, especially as the trends for the election outcome started becoming more and more evident. And if an administration that had high degree of acceptance for Bitcoin would come in place, that Bitcoin price would appreciate. And so you have to look at what's the best use of your capital in that moment. Can you get a great return by buying Bitcoin at a lower price now and then being able to leverage it over the long term to generate yield or you're going to invest that in an asset where the first Bitcoin that comes from that asset won't be produced for at least 12 or 18 months. And so we chose to leverage converts as a way to buy Bitcoin and you take an appreciating asset and use that as the basis for those bonds, which is better than using converts necessarily for buying depreciating assets such as miners. Back in late 2021, we did a convert where we raised $850 million, which we invested in mining infrastructure. And we learned a lot during that process and we were successful in buying back a fair amount of that at a discount, which created some great value for our shareholders. But we believe that convert debt is best used for buying appreciating assets, and that's certainly what Michael Saylor has done and been successful with it. And we'll see if it continues to be. He continues to buy a lot of Bitcoin and we're more opportunistic with how we view that, so...

And just to add to that, Brett, this is a capital allocation question. And as you can imagine as a large -- largest player in this space as a public company, we have multiple initiatives and we allocate capital to the best use of that dollar. And it made sense for us to go out and invest in Bitcoin. And you look at the history, we've created, almost 63% yield on that on a per share basis, which we're very happy about. Now in terms of our sources of capital, as you can see, we've been diversifying our source of capital. For example, last year, we -- the year before that actually in 2023, we had almost 100% reliance on ATMs. And last year, we reduced that to under 45%, approximately 43% or so. And the rest of the capital was sourced towards from the converts and also loan against Bitcoin. And that we expect will give us a return better than sitting -- that cash sitting on the balance sheet or from a treasury management standpoint.

No, certainly, I agree. For this year, I think in 2024, you guys grew your installed hash by 115%. How should we be thinking about hash rate growth this year? I know your hash -- network hash pulled back a little bit over the last couple of weeks, but it also has hit above 1,000 a few weeks ago. Do you envision maintaining that, call it, 5.6% share you had in the fourth quarter or improving that as we progress throughout the year, and would you want to give any targets to hash rate for 2025?

I think I'm not going to give a hash rate target per se. I think the global hash rate is going to grow very dependent on the profitability of mining Bitcoin. One thing we saw recently is essentially Bitmain creating US publicly-traded large-scale miners like this company, Cango who overnight had 32 exahash of capacity because Bitmain essentially wanted to deal with them to take their older machines and have them take them over. And I think what we're seeing is now, especially in China, there is so much excess power in China and the cities in China have now been essentially given what appears to be the green light to use that excess power to mine Bitcoin as a way to generate revenues for those cities due to the condition of the economy. And so I think you're starting to see Chinese miners wanting to raise capital quickly to be able to begin mining in China again. So there are some people who believe that global hash rate will go to somewhere around 1.2 petahash or 1,200 exahash by the end of this year. If Bitcoin price remains at its current levels, where it is today, I don't think we'll see that type of growth. I think we'll see growth actually potentially slow down and maybe even stop on kind of that large scale. So our intention is to continue to grow, and we're going to grow in an opportunistic manner. We don't believe that it is prudent for us to grow no matter what the cost. We think that the -- right now, the marketplace for compute for Bitcoin miners is still very attractive from a price perspective. But we also have the advantage that this company that we co-founded a few years ago, Auradine, has very successfully…

I think that's very well said. On the -- I guess the energy front plus maybe geopolitical front, do you think there's something that this administration or even future administrations can do to help make the industry more competitive when it comes to energy costs, given miners do help balance the load, they are a net benefit to the grid. Do you think there's something that could be done to ultimately help lower the energy costs for miners?

I think what you're going to see is, you have two challenges in the US. One is transmission capacity and the other is permitting of additional capacity. I think if you remove the limitations from permitting of additional capacity and you allow Bitcoin miners and AI operators to operate behind the meter freely, then what you'll see is the commercial markets -- the private markets and commercial markets go in and do upgrades to the grid in this country, which the public utilities can't do. And I think it's -- I'm not saying privatize energy, but what I think that you'll see this administration being focused on doing is how do we remove the friction in increasing energy generation, transmission, and making consumption more efficient. And that's all you need to do to make Bitcoin miners more successful. We have today, if you look at global hash rate, publicly traded miners make up roughly 35% of global hash rate, and the US arguably has about 40% of the global hash rates within its borders. The US could have much more if energy generators had an incentive to work with them. And that's a business decision, that's not the government mandating anything. But if the government would remove the friction from building new capacity from Bitcoin miners acquiring capacity, operating behind the meter, same thing for the AI operators, I think you would see things take off because what is good for the AI industry in regards to energy is very good for Bitcoin.

Perfect. I think that's a great place to leave it. Rob, may be back to you for some of the retail questions.

Thanks, Brett. Yes, we really appreciate all the questions. Let's turn now to some questions from our retail investors. The first question we got was, Fred, with the increasing competition in the crypto mining sector and potential regulatory changes in the US and abroad, what strategies is MARA implementing to sustain profitability and ensure long-term growth? And what is the end goal for us? Are we going to be a store value company like MicroStrategy or strategy or continue to mine Bitcoin?

If you think about MARA, we are an energy transformation company. We're in the business of turning energy into digital value. So we call it digital energy. And if you look across the spectrum of what we're doing is we're coming more and more vertically integrated from energy all the way to the holding of the actual asset and maximizing every piece of the value stack, if you would, and maximizing every share of wallet in the process that we can by becoming one of the top operators. Bitcoin mining is a zero-sum game. If you're the most efficient operator, the lowest cost operator, you will always be able to operate regardless of cost unless there is a major crisis. But you will always be able to operate. And when you have those types of cash flows, if you're an efficient deployer of capital, you can use your capital wisely, then you can grow and actually build significant shareholder value with less and less dilution over time. And I think the Bitcoin mining business needs to become it's more capital efficient and really the business needs to mature into the scale that the large miners have received. I think there's going to be a lot of consolidation over the coming years as most of the small miners just won't be able to operate efficiently. Where we're really headed is in leveraging the low-cost energy that we're in the process of acquiring or uses in the AI world and to develop technologies that support inference AI. Our liquid cooling technology is one of the best in the world, especially for inference at the edge. And I think if you were to see some of the solutions using our 2PIC technology for small scale inference at the edge, you'd realize that you now make deploying modular small data centers very easy where you can put them inside buildings where you don't need these massive cores and things like that and they generate no noise. And so we are becoming more and more a technology company and more and more of a solutions company. And I believe you'll start seeing us becoming more tightly integrated with data center operators at both a hardware level, a software level, and a services level. And we'll continue to grow Bitcoin mining and that technology together, and they'll provide a very well-balanced business.

Terrific. Our next question, Fred and Salman, that we received is, will there be any potential government partnerships or sponsorships in the near-future.

I think what you'll see is, in Bitcoin mining specifically, government partnerships will be more of an international occurrence. I think domestically, it will be with utilities and the actual power generators. I think what the government really wants to do is essentially try and remove the friction for Bitcoin miners and AI developers and they'll start putting together programs to do that across the country, I think, especially as it relates to power generation. And we're super excited about that. Now, as I mentioned earlier, 24 states in the US have said they're going to -- that they have Bitcoin strategic reserve measures or initiatives on their ballots. There's an initiative in the Senate to do the same thing for the US. Other countries are starting to look at that. And so then you start looking at, do governments want to go into partnerships with large holders of Bitcoin and do things. That's a totally separate area and I think that's an area that will be explored over the next two years most probably.

Great. And our last question ties into that one. What role does MARA expect to have with the Trump administration and their stated intention to ensure that Bitcoin is made in America? are we working directly with the administration?

Well, I think all of the large Bitcoin miners are working with the administration in some shape, form, or manner. We're very focused, obviously, in driving support for Auradine miners made in USA. We're very focused on mining blocks in USA. We stamp every block that MARA pool makes with made in USA. And we're trying to really build out the block space within the US because we believe it's of strategic importance to the United States, not just having a strategic reserve of Bitcoin, but also having a predominant amount of block space such that the US can project power and defend the Bitcoin blockchain regardless of what other nations do. So we'll collaborate with the government both the executive branch and the legislative branch at the federal level, at the state level to make sure that what's best for the industry and gets put forth and executed.

Terrific. Well, that's all the time we have for today. Thanks, everyone for joining us. If you have any questions that were not answered during today's call, please feel free to contact our Investor Relations team at ir@mara.com. And with that, I'll turn it back to you, operator.

Thank you. And with that, we conclude today's conference call. All parties may now disconnect. Have a great day. Thank you.