Welcome, everyone. I'm Jean with Three Part Advisors, and I'm delighted to welcome Kyle Turlington with Atlas Energy. Go ahead, Kyle.
All right. Good morning. Thank you guys for hearing our story here. Again, I do investor relations for Atlas Energy. It's going to be an interesting little 35 minutes here. We've got a lot to talk about. Real quick, you can't tell the Atlas story unless you know a little bit about our founder, Bud Brigham. I think he is a very important part of the story. Bud is infamous in the oil and gas space. Bud has enriched the lives of many people throughout his career. Bud is known for his first company, Brigham Exploration, which basically, Bud was one of the early wildcatters and used 3D seismic for horizontal drilling, basically discovered the core of the Bakken. He started that company with $30,000 of family investments and exited, went public, and I think in 2001, sold to Statoil in 2007 for a very incredible exit.
Bud then started a company called Brigham Resources, which basically proved out the southern Midland and southern Delaware Basin in the Permian. That company was sold to Diamondback in 2015. Bud then started a minerals company, which went public shortly, I think around that same time. Brigham Minerals was acquired by Sitio Royalties, and Sitio, I think, was acquired by Viper. The story for Atlas begins kind of after Bud's exit with Brigham Resources. At that time, there was a kind of a land rush happening in West Texas for local sand. We quickly discovered around that time period that we did not need the Northern White sand. It basically, the 100 mesh sand, operators became very comfortable with that, and we obviously, if you've ever been out in West Texas, there's a lot of sand deposits out there.
There became a land rush around 2015, 2016 to move a lot of the mines to West Texas from up north, and Atlas was formed at that point. Bud, being an E&P operator, always knew that having the best rock always made a big difference. Bud basically found two sand reserves he thought had the best geology for production of sand. Atlas was formed in 2017 as a new entrant into the market. Atlas has a couple of very distinct advantages versus the competition there. First off, because of the geology and the reserves we sit on, we have very low mining costs. So we are several dollars a ton cheaper than our next closest competitor in the production of sand. The other thing is Bud's always been an innovator.
Bud realized that the way sand gets delivered in West Texas was always kind of archaic. The way, and it is still primarily used today, is we bring in what we call third-party trucks to pick up sand at the well sites, and then the average sand delivery is around 70 mi-80 mi. in 2019, 2020, Bud had the idea of let us find a way to become more efficient in the delivery of sand, and the idea of building a conveyor belt, of all the ideas he had, the one that kind of made the most sense. They spent two or three years trying to get the right of ways. In 2022, they decided to move forward with the project and went public in 2023 to fund the construction of the Dune Express, which is the largest now conveyor belt in North America.
That conveyor belt is up and running today. It has been operating quite well. Today, Atlas is the largest sand and logistics company in West Texas. We have got about a third of the market share. We are going to produce between 23 million and 24 million tons, which is about a third of the market today. We deliver all of our sand. Not all of it goes down the Dune Express, because the Dune Express only covers the Delaware Basin, but probably about 35%-40% of all of our sand now gets delivered down the Dune Express. The legacy business is, I call sand the legacy business because we have obviously been pivoting into power, and that is going to be where a lot of your questions are going to come from, is my assumption.
After the completion of the Dune Express, Atlas had kind of built, put a lot of capital into a lot of these businesses. The last couple years, as many of you know, oil service pricing in West Texas, despite the elevated commodity price, has been kind of at trough levels. We do expect some sort of uplift in activity next year. We may or may not get that, because you guys have done a good job of hammering home capital discipline to a lot of these operators. It certainly seems like where the commodity price is today, that you will see an uptick in activity. Whether or not we get that or not, I do not know.
We generate a lot of cash even at the kind of trough levels for that business because of the advantage mining costs that we have, and then obviously the outsized margins we make on sand delivery. Back as we were kind of coming toward the end of this capital cycle for the sand business, as the Dune Express construction was being finalized, management teams looked for ways to continue to grow. We kind of felt like at the time that there was just not many more opportunities in the legacy sand business to really grow. At that time, talking to a lot of our E&P customers, one of their biggest concerns, this is kind of back in 2024, was power, because the grid, obviously, in West Texas was becoming a big problem.
Operators were having a very difficult time finding power for a lot of their production, a lot of their completion activities, there just wasn't a lot of grid access out there. In early 2025, Atlas bought a distributed power business called Moser Energy Systems, which provides distributed power for the oil and gas space. That has been a really strong acquisition for us. We've basically, over the last 18 months, have basically got that business to where we basically are out of equipment now. There continues to be a really stable, growing demand to provide power for the oil and gas space. That business was kind of our entrΓ©e or sort of appetizer into what we felt like was opportunity to change the complete cash flow profile of the company.
Back in the fall of last year, 2025, we purchased 240 MW of recips from CAT. These are kind of 4 MW recips, which was our first big purchase into the behind-the-meter power for data centers as well as for commercial industrial opportunities. We were helping a customer at that time find equipment, and we purchased that equipment for that specific customer. We ended up in March 31st of last year contracting 120 MW for a five-year contract in El Paso for a C&I customer. We're providing permanent power for them, that customer. We've got multiple renewal options for that customer, and that project will be up and running in March of next year. So March 31st of 2027, that project will be up and running. We've already got all the gensets placed there, and construction there is going really well.
When we acquired that equipment, at that time, we were starting to have early conversations with data center customers for equipment. The problem that we had at the time was data center customers want surety of supply, and they want large amounts of equipment. So, coming to a data center with 120 MW of equipment doesn't really get you into the big boy room, so to speak. It became apparent that in order to execute on this opportunity, we needed more equipment, and we needed surety of supply. Back in March of this year, we entered into what we call a Global Framework Agreement with Caterpillar, where we agreed to purchase 1.4 GW of equipment with equipment deliveries in 2027, 2028, and 2029.
So, about 300 MW-400 MW a year for the next three years for CAT, and that put us in position to where we could obviously enter into providing permanent power for data center customers. Back in the second quarter on our Q2 call, we kind of let the street know that we had two projects planned for Texas for behind-the-meter deployment. We were fairly far along in the negotiations at that time. We were excited to have contracts for those projects in Texas by the end of the year. Lo and behold, and a lot of you guys in the room will know that the election results in Texas went a different direction than a lot of us probably thought or what we have seen historically in Texas.
After the results in Texas, a lot of people became really concerned that we've got these two projects in Texas. All of a sudden, it looks like data center development in Texas is going to be in limbo land for a period of time. We obviously got hit with that news. There's been a lot of concern around AI build-out at this time. Reality is that our customer desperately needs computing power, and with Texas being sort of uncertain at this time, we've been working with that customer to move outside of Texas until we expect to have that same sort of deployments that we were expecting in Texas. We expect to have those deployments in states outside of Texas. We can talk a lot about Texas and what we're seeing there on the data center development side, but we expect to pivot pretty quickly.
We expect to have that same sort of equipment deployed, but we now expect that equipment to be deployed outside the state of Texas. We've been looking at sites in Oklahoma, Mississippi, Louisiana, South Dakota. We've actually narrowed down to a specific site that we expect to be deploying this equipment for. I think it's a really good opportunity because the market is certainly spooked around data center growth, and yet the reality is, you look at some of the leading We were public and said our customer is one of the leading AI labs out there. They've got these huge mandates to get their own computing power, and with Texas in limbo. They had a lot of projects slated to go in Texas.
The need for them to pivot and get these projects online as quickly as possible is out there, and it's easier path to get these projects online earlier now in other jurisdictions. We expect to have news to share on that, we think, by the end of the year, which is a little bit later than what we originally thought. We thought we'd have these things wrapped up by now in Texas, but we didn't necessarily see that curveball coming that the political landscape gave us in Texas. We're excited. I think this is a great opportunity because the market is assuming that a lot of this data center growth is going to slow. It's going to slow in Texas. I don't think Texas is going to see a lot of projects start up this year.
I think Texas will still be the epicenter of data center development. I think we're going to have to see legislation passed, and I think it's going to be probably not till the spring till we really know what the rules of engagement are in Texas. I think some of the projects that were initially slated to go in 2027 are going to be moving elsewhere. I think Texas will, once again, see a lot of projects developed in the state, but I don't think some of our early projects now are going to be in Texas. We're excited about sharing some more news with you guys down the road. There's a lot in there, but that's kind of the quick update on Atlas.
I am really here to answer some questions, and I know probably there are some questions around Texas and what we think is happening there. I was going to leave the room up for Q&A at that point, but that is the quick and dirty on where we are, and we are really excited about this pivot to other jurisdictions. Again, we hope to have news to share with you guys before the end of the year on some of these projects. A couple things real quick I was just going to mention. With Texas obviously having some issues, there are a lot of opportunities. There are a lot of states that are welcoming for data center development. I think that the four things you need to look for a site is, number one, is there community engagement? Is there community buy-in for that project?
If you look at sort of, I know there is a lot of backlash out there in some states. There are some states that desperately need this revenue, and some of the tax revenue you can bring into some of these counties just dwarfs what their existing tax revenue is. You are talking 20x potential tax revenue that you could be bringing to these communities. So, believe it or not, there are some red states out there are some communities that desperately want this. They want the jobs, they want the revenue. The first thing you really need is community buy-in. The second thing you need is you need access to gas. A lot of the states we are looking at, Oklahoma, Mississippi, Louisiana, it is not very difficult to get gas brought to your site.
When we look to go outside of Texas, that is one of the. We wanted to see the community buy-in. We want to see that there is access to gas, and some of the sites we have selected, the site we have selected obviously checks that box. The third thing you need is what they call a sleeving arrangement. You need to have the local utility. You need to have a very favorable setup there. Some utilities are a little bit more difficult to deal with. If you have got a co-op situation, those tend to be a little bit easier to deal with. So, negotiating your sleeving arrangement with utility is something that you need to make sure that it is. Some are easier than others. So, you obviously want to have this. You want to make sure that that is something you think is an easy negotiation.
The fourth thing you really need is the permitting process. You need to have clear line of sight on the permitting side and pretty straightforward regulations in place in that state. Some of the states I mentioned before have got a clear line of sight. Ironically, Texas had that in place until very recently. Texas was probably going to see 60%-70% of all data center developments were probably going to happen in Texas, and I think now you are going to. That is certainly up in the air for the period of time. But we are really excited about putting our assets to work in other jurisdictions. Believe it or not, despite all the headlines out there, you look at the need for compute that is out there.
You look at companies that there's almost a quarter of a trillion dollars in chips that have already been purchased that are sitting in warehouses that have to be deployed. You've got companies that their entire business model is based on getting their own internal compute. There is a real sense of urgency from our customers to get these data centers built. Again, I think everybody's been too focused on Texas. There's a lot of other sites out there where I think these things can get rapidly deployed. I think Texas is going to be fine. I just don't think Texas is going to. Until we get the new rules and regulations from the new governor, it's just going to be hard for Texas projects to kick off the ground anytime soon. Just a little bit more perspective there.
But again, happy to go any other directions or answer any questions you guys might have.
Could you address the balance sheet and the bridge loan and expectations around that?
Yeah, that. Perfect. So we entered into two CRAs, which is called a cost reimbursement agreement with our customer. The reason behind that is because we, at the time, we needed to, as we were. By the way, one of the blessings in disguise here is that we did not sign an ESA for some of the Texas projects, because had we done that, I think we would have been kind of tied to those sites at that point, and we'd have to wait through, we'd have to wait things out in Texas. Again, I think we'd have been fine at that point, but the fact that we did not sign those ESAs allows us to pivot and pivot pretty quickly. But, yes, so we entered into two cost reimbursement agreements with our customer. Why did we do that?
This was sort of in October, and as we were waiting for the final contracts to get signed, we needed to order equipment for them. There were some fairly long lead time items for balance of plant equipment that you need to secure. In order to do that, we had to place an order there, and as we were waiting for the ESAs to get done, we pivoted to what we call a CRA, where the customer sort of guarantees the financing of that equipment. We entered into two CRAs. One's for balance of plant equipment, one is for incremental equipment that the customer wanted that was available in the market from a project startup in Utah that was being delayed.
CAT came to us with incremental equipment because the size of these projects we're working on are quite large, and we needed incremental equipment. We then entered into a bridge loan facility with Goldman Sachs and Barclays to fund the purchase of that equipment, and that bridge loan is backed by the CRA. In the event that we do not come to a contract discussion or we don't get to a contract with them, all the liabilities of the bridge and the purchases we've made fall back to the data center customer, and we get reimbursed for any purchases we made. That is sort of our protection around if we don't get a contract signed. We do expect to have a contract signed here by the end of the year.
Once we do, we'll go to the markets to get permanent financing on those projects, and then that permanent financing will be used to pay back the bridge facility that we currently have with Goldman Sachs. I don't know if that completely answered your question or not. We do have purchase obligations this year with CAT for the first tranche of the GFA, the 27 equipment, about $270 million. I think you can expect us to get that equipment added to the CRA here very soon, which will cover any of the near-term liquidity concerns that are out there.
There's a fear out there that because there's delays in a lot of these data center contracts with our financing requirements, there's concerns around the balance sheet, and we will alleviate those concerns here very soon. The plan is to put the CAT equipment that we are purchasing this year into that CRA, and we'll get additional financing through the bridge loan to cover that equipment. The plan obviously is to have the ESA signed by the end of the year, and then permanent financing will hopefully be secured sometime in February or March, and at that point, the bridge will get repaid, if that makes sense. The existing business, the legacy business is generating a lot of cash. It's impressive.
One of the cool things about Atlas is that even with trough conditions out there in West Texas, the base business generates around $10 million of EBITDA a month, or cash flow a month, I should say. If we are able to get better pricing in that business, that business can spit off a ton of cash. If we got sand pricing started to recover every dollar increase in sand pricing, if we're producing 25 million tons, that's $25 million of incremental EBITDA that falls straight to the bottom line. We are a coiled spring certainly for improvement and activity in West Texas, but I think it's good to know that even in a trough conditions, we can generate a lot of cash. One of the other catalysts, I think, for Atlas is logistics pricing. We deliver all the sand that we produce.
Logistics pricing last year was very depressed. Obviously, thinking back to last year, we had lower activity. We also had a surplus of truck drivers. The over-the-road market was very weak, so a lot of drivers were forced to work in West Texas, because pricing on the over-the-road market was pretty depressed. Fast-forward to today, the over-the-road market has gotten quite strong. It's pulled a lot of drivers out of West Texas. The early phase of data center development is starting to pull drivers away from sand and logistics. They make almost twice so much money per load working for data center customers as they do for sand deliveries. We've had some immigration crackdown on foreign driver's license, so it's removed some drivers from the pool. Logistics is actually and then obviously, higher diesel prices is also out there.
We think that there is a very strong tailwind for higher logistics pricing. That is very advantageous to us because of the Dune Express. We want trucking rates to move higher. West Texas pricing's still below over-the-road market, so we feel like we've got a big opportunity, even without an increase in activity, to continue to see higher logistics pricing. Our margins in logistics, we guided to mid-teens margins in the fourth quarter. We think they have potential to continue to grind higher next year. Over time, I think that we could get to kind of high teens, low 20% logistics margins in logistics. That should be a nice tailwind. Again, I don't think that we necessarily need to see an uptick in activity for that to play out, but we do need to see an uptick in activity to see sand pricing move higher.
There is a hope out there that, as we've been in trough conditions for sand pricing, and considering the cost curve out there, a lot of our competition has been cash flow breakeven or even losing money. We continue to hope to see more mine closures. It's just really hard for guys to throw in the towel when they see where the current commodity strip is out there. Hope is a powerful thing for a lot of folks. We do think that there is a chance that if guys are tired of losing money, a lot of our competition is small mom and pops, with additional mine closures, we could see the market tighten up with that. Again, I probably would've said the same thing to you guys last year, and we haven't seen a ton of mine closures.
That is something that we're certainly monitoring, and we know the health of our customers is bad. We're seeing that through the grapevine in West Texas. Mine closures is also something that could certainly reset supply and demand out there for sand pricing as well. I'll touch on the permitting situation in Texas and why I think things are going to be kind of tough to move forward. You guys might remember back in August, Governor Abbott put a moratorium on data center development, or asked for an audit for a data center development that was seeking grid connectivity. That was actually a really positive for us because we're providing a complete behind-the-meter solution, so we're building a private grid for our customers.
That created a big tailwind and created more of a sense of urgency from our customer and then really grew our pipeline of opportunities. The curveball we were not expecting is that in September, Governor Abbott then came back and put a moratorium on any sort of air permits in Texas. We were not expecting to see a moratorium from a Republican governor of Texas as it relates to the air permitting situation. We had a permit filed in Shackelford County. Ironically, that air permit has actually been approved now. The problem that we have in Texas is that when the governor takes over in January, one of her mandates, one of her four pillars of things she was running on was to slow down or put a moratorium on data center development. Look, I think she wants it done responsibly.
I think she does want this revenue and these jobs to come to Texas. The problem that we have is that until we get the new rules and regulations, which the legislature will meet in the spring, we think that one of the first items they'll take up is data center reform. I do believe that Texas will pass legislation that will keep Texas as sort of the epicenter of data center development. The problem is we won't know that until probably late spring, early summer. Those rules will probably take effect in September, which makes it really difficult to launch any sort of new projects. Projects that have already been permitted and construction has started, I think, will continue to move forward. Otherwise, it's difficult to, even if you were able to secure your permits, for the credit markets to underwrite anything for you.
One of the things that we will do once we get our ESA signed, our contract signed, is we'll go to the credit markets for the funding of those projects. Texas seems like without sort of clarity on that, it's going to be very difficult for the credit markets to underwrite our projects. Which is why our customer and us are pivoting to another jurisdiction, because you just don't have that worry. The permitting process in other states is pretty clean and straightforward. The one thing that credit markets don't want is uncertainty. Providing a project that doesn't have any sort of risk associated with it is why we think moving to another jurisdiction makes the most sense. Just wanted to provide a little bit of a background there as well.
So, at the end of the day, assuming this goes forward as planned, your balance sheet is still going to contain quite a bit of debt. Is that correct?
We will have. Yeah. These projects are going to be large. So, Chris, you're correct about that. The projects will, look, we've been very public talking about the cost of these projects being somewhere around $2 million-$2.5 million per megawatt. So, if we're looking at a project that's 500 MW, let's just use a gigawatt project, for example, the project cost will be around $2 billion-$2.5 billion for the cost of that project. Now, we do have some equity already in place for a piece of those projects, but there will be some outside equity that we will have to bring in for these projects. The cash flows from these projects are going to be pretty strong as well. We've been very adamant about the economics of these projects. They would be 15- to 20-year projects, EBITDA paybacks in the five- to six-year range.
If you're looking at a project that's going to cost $2 billion-$2.5 billion, you're going to be looking at EBITDA generation on an annualized basis north of $400 million. So, yes, there'll be a lot of debt. We think that the public markets will fund about 80%-90% of that capital. We'll have some equity we already have in place for these projects. Where our stock price is today, there is no desire to issue equity at the parent. We will bring in whatever additional equity we need. We will bring in sort of project-level equity for those projects, so more or less selling a piece of the project to an equity investor. So, there should be no fear around a big dilutive event happening for us once we announce these projects.
But yes, if you look at the current debt we have on the We have a $540 million term loan. That term loan basically runs the sand business. That sand business is generating decent cash flow. That term loan is being paid down. We have a $450 million convert, which part of that is funding the Socorro project, which that project cost is $190 million. It's going to generate $50 million-$55 million of annualized cash flow. And then the remainder of that is equity we're going to be putting forward to the first project. The bridge loan is all under CRAs. That is on our books today, but that is covered by our data center customer.
To the extent that we don't get to a contract, we will, but to the extent that we don't, that bridge loan, we get reimbursed for anything that we have spent, and then any obligations that are left on that fall to the data center customer. In a way, we're covered on that, if that makes sense. Does that answer your question?
Yeah. Then, assuming things go to plan, it'd be 2028 before you start receiving revenues off of this?
Our plan is to start construction on these projects in February, which means we've got to go. Our customer needs power by the end of 2027. Those are the timelines we're trying to hit. Very likely, these projects could start up in early 2028 instead, because we are obviously coming up on those timelines. But yes, these projects will certainly, we think, late 2027, early 2028, you'll see first power. They'll probably ramp, because we will not have all that equipment delivered on day one. But that project will likely ramp through the middle of 2028, and by the middle of 2028, that project should be fully up and running with the full amount of power. But we'll provide all that detail when we have the contracts to announce. Does that answer your question?
Yeah.
Yeah.
How much wiggle room do you have with the purchase commitments regarding CAT, and have they expressed any interest or capability of offering you some breathing room given what's going on in Texas?
Well, we've got a really good relationship with CAT, and our customer is probably CAT's most important customer they have. They obviously I think the fact that 2027 equipment is so valuable, the fact that CAT gave us 280 MW of incremental equipment a couple of months ago, or a couple of weeks ago, is a good sign that CAT knows who our customer is. CAT knows what our customer's needs are. The fact that CAT gave us that equipment, again, it's under a CRA, so the obligation is going to fall to our customer. Look, I think that you're going to, we think that all of our 2027 equipment obligations to CAT will get put into a CRA here very soon, and that will alleviate any of those financing concerns, because that will be covered by the CRA.
But I think CAT is realizing that it's not just us. I mean, there's a lot of projects that they were expecting to get deployed in Texas. Some of these are actually in a tough spot because we just don't know what's going to happen in Texas. If you've already got a project committed, how long do you have to wait before you are able to start construction? As I said earlier, it's sort of a blessing in disguise that we did not get a contract signed in Texas, because we'd have been kind of very likely tied to that. But I think CAT is working, and I think this is a great opportunity for us, because I know what the computing needs are for our customer, how much they need to add.
With a lot of projects in Texas uncertain, we think there's a good chance we can grab significantly more equipment with CAT to deploy quite a bit more equipment with our customer, because they're desperate for I mean, their entire business model is based on getting more of their own compute. It could be a real advantage for us. But yes, CAT is going to work with all their providers, and again, they know who our customer is. It is CAT's most important end customer, and they'll certainly work with us, but I don't think we're going to need to, because I think we'll get that. All of our 2027 equipment will get put in a CRA, and therefore take that financing risk off the table, if that makes sense.
Got you. Your company has been very nimble in finding new opportunities. Is there anything else that you're working on besides these two main lines of business?
No. Right now, obviously, think of the sand business and even the oil and gas power business. Right now, we're kind of running them in kind of harvest mode, so minimal CapEx, almost no growth CapEx in that business. There will be a little bit of growth CapEx tied to the oil and gas side, for projects that we'll build equipment for projects that we can get sort of long-term contracts in. Long-term contracts in the oil and gas space is like three years, three to five years. I think we're very happy with being almost completely sold out of that equipment, and kind of harvesting that for cash. The sand business, we're in harvest mode for that as well. A lot of our focus right now is connecting on these behind-the-meter projects.
Again, I think we'll have news to share with you guys very soon on all that. Those projects can completely transform the cash flows of the company, and we could not be more excited about the opportunities we have there. We have time?
Thank you.
Okay, thank you. All right, appreciate it.