MP Materials Corp. (MP) Earnings Call Transcript
February 26, 2026
Earnings Call Speaker Segments
Thanks, operator, and good afternoon to all of you. It's a pleasure to welcome everyone to the MP Materials Fourth Quarter 2025 Earnings Conference Call. Joining me on this call from MP Materials are Jim Litinsky, our Founder, Chairman and Chief Executive Officer; Michael Rosenthal, our Founder and Chief Operating Officer; and Ryan Corbett, our Chief Financial Officer. Just a reminder that during today's discussion, we'll be making forward-looking statements about future events and expectations, and those statements are subject to a variety of assumptions and caveats. Factors that could lead the company's actual results to differ materially from those statements can be found in today's presentation, the earnings release, and in our SEC filings. Additionally, this presentation includes some non-GAAP financial measures. You can find reconciliations to the most directly comparable GAAP financial measures in today's earnings release and the appendix to today's slide presentation. Whenever we refer to EBITDA in our discussion today, we mean adjusted EBITDA, and when we say tons, we mean metric tons. And finally, both the earnings release and the slide presentation can be accessed on our website; so with that, I'll hand the call over to Jim. Jim?
Thanks, Martin, and I appreciate all of you being here this afternoon. For MP Materials, 2025 stood out as a defining year — from the transformative partnerships we formed with the Department of War and Apple, to faster expansion throughout our operations, to manufacturing magnets at Independence using equipment at commercial scale. We're finding our rhythm as America's champion and as a vertically integrated worldwide leader in rare earth magnetics. Now let's move to Slide 4. I'll begin with the Materials segment and what's happening at Mountain Pass. During 2025, our NdPr oxide production doubled, reaching 2,599 metric tons. What matters greatly is that we closed out the year running at an annualized rate of close to 4,000 metric tons of separated NdPr oxide. Michael will go into the newest operational accomplishments and what Q1 is showing us in greater detail shortly. With output climbing, total oxide sales volumes increased 75%, approaching 2,000 metric tons across the year. This production and sales ramp arrives at a particularly advantageous moment, since the price protection agreement has taken effect and NdPr pricing has also risen considerably in recent weeks. Engagement with customers is getting substantially stronger, and I'll return to that shortly. We likewise surpassed 50,000 metric tons of REO produced during 2025 — yet another record annual result. In our view, we're still the second biggest producer of total REO worldwide, which highlights the outstanding quality of the Mountain Pass asset. While upstream and midstream output grows, demand for NdPr oxide stays strong. That strength was underscored by the recent signing of a major long-term NdPr offtake agreement with a new strategic customer, one of America's leading technology and industrial companies. Today we hold direct strategic agreements with 4 of the world's leading manufacturers spanning automotive, consumer electronics and physical AI; through these partnerships we can scale our NdPr business now, with an obvious route toward downstream magnetics growth as time goes on. Work continues on our heavy rare earth separation circuit, and we remain on schedule to start commissioning at midyear. Consequently, we anticipate producing separated heavy rare earths — dysprosium and terbium specifically — in late 2025. Last, within the Materials segment, better realized prices, PPA benefits and ongoing cost reductions produced a solid return to profitability, delivering $40.3 million of adjusted segment EBITDA in the quarter. Now to Magnetics. We met our goal of producing our first magnets on commercial scale equipment at our Independence facility late last year, and we're currently optimizing the startup of the production systems. Qualification of the commercial processes with our foundational customer has begun, and we anticipate initial deliveries and revenue during the second half of the year. Meanwhile, expansion of both recycling and magnet capacity is proceeding, which generated an additional $32 million prepayment from Apple, earned in the fourth quarter and received earlier this month. Another important milestone this month was choosing Northlake, Texas as the location for our new 10X facility; we obtained more than $200 million in incentives and grants to back the build-out. We anticipate breaking ground very soon, with engineering work and procurement of long lead equipment already progressing substantially; consistent with the DOW, we're moving this project forward quickly and with disciplined execution. Ryan is going to walk through the investment profile we expect for 2026 as we speed up the 10X build-out. Our output of precursor magnetics products, NdPr metal in particular, kept moving ahead. We achieved record quarterly volumes for both production and sales while also improving yields, which allowed us to maintain robust margins. Consequently, the Magnetics segment delivered $8.4 million in adjusted EBITDA for the quarter, which brought Magnetics EBITDA for the full year to $26.4 million. Behind the scenes, the Magnetics team keeps performing at a very high level, not just in optimizing processes but also in creating proprietary intellectual property. As one illustration, our technical team has made substantial progress in grain boundary diffusion capabilities, producing a magnet formulation and production process that consumes roughly 60% less heavy rare earth content than we had initially expected, while still satisfying the stringent specifications needed for high-performance, high-temperature, EV-grade magnets. Since our initial assumptions were already at the low end of industry norms, this amounts to a significant technical accomplishment. Reductions of that size demonstrate the strength of our engineering bench and how disciplined we are in execution. They also strengthen our competitive position. Heavy rare earth elements like dysprosium and terbium mainly boost coercivity, which improves resistance to demagnetization at elevated temperatures; they don't raise magnetic strength or torque density, and at higher concentrations they can even lower energy product. For many physical AI and humanoid robotics applications, operating temperatures are considerably lower than those in EV traction motors, so performance depends more on magnet design, light rare earth optimization and motor architecture than on heavy rare earth content. As magnet technology improves and adoption increases, especially in physical AI, performance and value creation are increasingly centered on NdPr. So we expect long-term demand growth and pricing strength for NdPr to greatly exceed that of dysprosium and terbium; with that, let me hand things over to Ryan. Ryan?
Thanks, Jim. Turning to Slide 5, you can see how we performed on a consolidated basis for the year. Looking at the left side, revenue grew 10% year over year. The main drivers were the ramp-up of oxide sales within the Materials segment, plus the first sales of precursor products in the Magnetics segment, which didn't exist back in 2024. We'd also like to call out what the Price Protection Agreement, or PPA income, contributed — that added $51 million during the quarter. When you combine that with reported revenue, what it really shows is that we realized the entire $110 per unit purchase price floor on products sold, and on top of that we got a strong contribution from stockpiled inventory, given where pricing stood during the quarter. Toward the center of that slide, adjusted EBITDA improved considerably over 2024, a shift powered by PPA income we earned plus sales of profitable magnetic precursor products. That growth in adjusted EBITDA served as the primary driver behind the year-over-year gain in adjusted diluted earnings per share. Moving on now to Slide 6. Here we present the Materials segment's annual operating metrics, with upstream key performance indicators over on the left side and midstream KPIs over on the right. Upstream results, as Jim noted earlier, were particularly robust — another record-setting year for production, exceeding 50,000 metric tons of REO in concentrate, which marks a 12% rise compared with 2024. Keep in mind that once third-party concentrate sales are removed around the middle of 2025, our reporting will stop including upstream sales volumes and realized pricing going forward. Over on the slide's right-hand section, NdPr oxide output keeps climbing, roughly doubling compared with a year earlier to hit about 2,599 metric tons. That higher output pushed NdPr oxide sales volumes up 75% versus the prior-year period, reaching slightly below 2,000 metric tons; now that the Price Protection Agreement is in place, we consider the realized price figure to carry less significance for investors. We do disclose NdPr oxide revenue alongside sales volumes, so anyone can easily derive an implied realized price by taking revenue and dividing it by volume. So going forward, realized price won't be shown as an official KPI. And then for the full year — if you look at Slide 7 — we break out financials by segment, with Materials on the left side and Magnetics on the right. Looking at Materials, revenue was down compared to last year, which mostly came from us stopping concentrate sales to outside parties. That drop was partly made up for by NdPr oxide volumes rising 75%. Even though reported revenue fell, PPA income plus better unit economics on NdPr oxide sales outweighed losing those concentrate sales. Because of that, 2025 segment adjusted EBITDA for Materials came in significantly higher than last year. Looking at the graphic's right-hand portion: once magnetic precursor product sales kicked off in early 2025, the Magnetics segment brought in $66.9 million in revenue for the year, along with $26.4 million in adjusted EBITDA. Now let's move to Slide 8, where we cover how we performed on a consolidated basis each quarter. Revenue declined slightly versus the same period last year, reflecting our decision to halt concentrate sales to external customers; stronger NdPr oxide sales and increased magnetic precursor product sales partially offset that drop. Just like on earlier slides, once you factor in PPA income, the combined number gives you a fair picture of the revenue we'd have booked at the $110 per kilogram floor price, plus what the stockpiled contained NdPr contributed; over in the middle of the slide, adjusted EBITDA rose considerably versus the same period last year, mostly on the strength of better NdPr economics. That same adjusted EBITDA gain was the biggest reason adjusted diluted earnings per share for the quarter went up. Let's move to Slide 9. Looking at operating metrics for the Materials segment, concentrate output came to 12,080 metric tons, which is about 600 metric tons above what we produced a year earlier. Keep in mind that, when you look at things quarter over quarter, the plant had scheduled maintenance turnarounds during Q4 in both 2024 and 2025, which is why those periods typically come in with weaker output than the preceding quarter. Turning to the right side of this slide, NdPr oxide production rose 74% from the same time last year, supported by continuing optimization and debottlenecking efforts at the plant. Volumes came in at about the same amount as Q3, again reflecting the lost operating time caused by the October shutdown, after which output recovered in line with what we expected. Michael will give you more detail on how we performed this quarter, plus what he expects for first quarter production and when he thinks we'll hit our target steady-state rate of roughly 1,500 metric tons per quarter. The progress our team has made on these debottlenecking efforts has us quite pleased. When we spoke last quarter, we noted a slightly longer-than-usual lag between making product and selling it, which basically works out to about 1 quarter of production sitting in the channel; so as production went up, that channel inventory went up right alongside it. That mirrors the ongoing metal production ramp in Southeast Asia, which includes intentionally building up on-site inventory needed so that various partners can run continuous 24-hour operations. Keep in mind that metallization happens at very high temperatures above 1,000 degree Celsius, and keeping electrolysis running without interruption matters a great deal for getting high yields and low unit costs. That makes stockpiling more inventory a required and deliberate part of scaling up efficiently. Still, going forward, we think that gap between production and sales will shrink and get a bit better, thanks to the continuing ramp of metallization capacity and the extra demand for oxide we've been seeing, including meaningful volumes from the new oxide contract Jim brought up earlier. As pricing gets better and our sales and metallization work keeps moving forward, we should see output turn into sales somewhat more quickly in Q1. A modest share of those sales will likely be intercompany, going to the Magnetics segment, so at the corporate level we'll have some revenue and EBITDA eliminations, similar to what happened in Q4. Moving now to Slide 10, this shows how each segment performed financially during the quarter. On the left, Materials revenue was down versus the same quarter last year, reflecting the fact that we've stopped selling concentrate to third parties. That decline was partially offset by higher volumes and improved pricing for NdPr oxide and metal products. Just like what we saw in the consolidated numbers, the main factor lifting Materials segment adjusted EBITDA both versus last year and versus the prior quarter was stronger NdPr unit economics, which PPA income supports. Turning to the right side of this slide, the Magnetics segment hit record levels for both production and sales volumes in the quarter. Even so, revenue and segment adjusted EBITDA were down a touch from the previous quarter, and that decrease is actually a sign of better yields and cost efficiencies, as Jim mentioned earlier. Before I pass things over to Michael, I wanted to walk you through a few things, beginning with pricing. Just to remind everyone, our contracts come in several distinct structures, and some of them tie pricing to trailing averages. Because of that, whenever market prices climb quickly, the upside doesn't show up right away in reported revenue — typically there's a delay of roughly 1 quarter to 1.5 quarters. One key point: for the NdPr products we sell, we usually earn the PPA off average contract pricing, so the gap between our contract pricing and the price floor will generally be reflected in the PPA income line in Q1.
One thing that's worth emphasizing, though, is that for NdPr sitting in stockpiles, PPA payments get figured using market prices with no delay. Consequently, once we move into the first quarter and market prices climbed rapidly toward the $110 per kilogram mark, we'll choose not to accept PPA payments on specific stockpiled NdPr materials. Keep in mind that taking PPA is entirely our choice, meaning that based on where pricing sits, we'll retain the ability at any time to keep concentrate and/or refined product inventory on hand and collect PPA payments on it down the road. Moving on to Magnetics. On the balance sheet you'll notice deferred revenue of roughly $74 million listed under current liabilities; our expectation is that this deferred revenue gets recognized across the coming 4 quarters at EBITDA margins roughly in line with those from the fourth quarter. As we get closer to first magnet deliveries during the second half of the year, we'll share more specifics regarding what we anticipate for added revenue and margin contribution coming out of magnet production. It goes without saying that we couldn't be happier about how far our Magnetics team has come in fulfilling what our customers require. Now to the balance sheet. The considerable strides we made in Q4 caused our other receivables balance to grow substantially, surpassing $131 million, which represents cash already collected in Q1 or due to arrive later in the year, among it more than $70 million owed by the U.S. government through tax credits and PPA payments, plus the $32 million progress payment tied to our work with Apple. On the investment side, now that we're starting construction on the 10X facility and pushing ahead with expanding Independence, recycling programs and heavy rare earth separations, we anticipate total capital expenditures landing between $500 million and $600 million in 2026, with the bulk of that going toward faster 10X investment and other growth efforts. Thanks to our solid liquidity, holding over $1.8 billion in cash along with future operating cash flow, we think we've got plenty of capital to carry out these initiatives. Growth in operating cash flow should come from recent pricing gains, the protection against downside that the PPA offers, continuing expansion of NdPr oxide sales and steady gains in unit costs. Taken together, this puts us in a good spot to completely finance these major value-enhancing projects while keeping our balance sheet strong. Last, because of the extra complexity that comes with the Department of War contract, we'd urge investors to look over our Form 10-K filed today, especially the footnotes to the financial statements. With that, I'll hand things over to Michael. Michael?
Thanks, Ryan. If you look at Slide 11, you'll see a number of the key intermediate product lines at Mountain Pass and Independence that feed into our finished magnets. Now on to operations. Q4 wrapped up an extraordinary year, and we're very pleased with how things went, since both Mountain Pass and Independence hit major production milestones. At Mountain Pass, we posted a solid quarter for concentrate output and yet another strong quarter for NdPr oxide output. At Independence, as planned, we started making magnets at commercial scale, using raw material that came from mining and processing at Mountain Pass, then was metallized and converted into magnets in Texas. The scheduled October outage weighed on our upstream and midstream operations. As I flagged on the previous call, some extra required maintenance plus a certain amount of rework meant a longer path back to normal production, and during the latter half of October more product was routed to concentrate rather than the refinery; even so, Mountain Pass finished the quarter carrying strong momentum, as December set a record for NdPr oxide output at an annualized run rate of almost 4,000 metric tons. Uptime keeps improving across nearly every circuit, and our focus is on raising throughput; along the way, we've found bottlenecks — some anticipated, some not — and we're working on them. Nothing is a showstopper, and we anticipate clearing these issues in a steady fashion across 2026. Let me restate what I said a quarter ago. For Q1, I'm looking for sequential production growth above 20%, then somewhat slower sequential growth across the following 2 quarters, with a reacceleration later in the year, so that we leave 2026 nearing our goal of 500 tons per month of NdPr oxide output, equivalent to a 6,000 metric ton annualized run rate. As always, the exact cadence may shift depending on when scheduled maintenance outages, project tie-ins and other upgrades take place. Meanwhile, Mountain Pass keeps getting ready for growth ahead. Work on our heavy rare earth separation circuits sped up dramatically during the quarter. What's left is mostly piping, electrical instrumentation and process automation work. We're still on track to begin commissioning and charging those circuits mid-year; in addition, we're finishing restoration and improvements at our idled chlor-alkali facility, where the first train should start commissioning during the second quarter. Pre-commissioning and checkouts are already underway on portions of the brine treatment systems. Since the agreement was signed in July 2025, the Mountain Pass team has engaged with Apple on the recycling initiative in a highly productive and interactive way. Process design is now final, and we're eager to move into the next stage of engineering. During the fourth quarter, our Independence team expanded quickly while substantially strengthening its technical depth; we connected the poles of our integrated magnet offering by commissioning the full oxide to finish magnet production process. Today we're turning out on-spec, high-grade magnets on commercial scale equipment, using oxides made at Mountain Pass. Let me be upfront: the coming year will be lengthy and full of obstacles as we tackle fixes, fine-tuning, more fixes, production scale-up, and magnet output growth. Even so, our beginning has been excellent. My pride in this team is enormous. In Q4, we moved quickly on purchasing and engineering work tied to enlarging the Independence site in service of our Apple collaboration. And, as Jim and Ryan noted, the groups handling capital projects, operations, and engineering are likewise driving 10X forward on the ambitious timeline we set out; during Q4, both the amount and the quality of alloy flake output rose sharply. We additionally brought online more equipment for powder output, pressing, sintering, GBD, and finishing. The commissioning work needed to kick off the formal production part approval process — PPAP — qualification with GM is now being wrapped up. Our expectation is that this allows commercial magnet sales to start in this year's second half. Though the physical process gets locked, our advancement and enhancements will not stop. Simultaneously, we are engaged with a growing roster of prospective clients regarding prototype grades and magnets across both Independence and 10X. I want to express gratitude to our teams for the tireless, steadfast dedication they showed throughout the prior year. Given the base and the momentum now in place, I eagerly anticipate yet another year of outstanding execution along with fresh milestones in 2026. That said, I will hand things back to Jim.
Thanks, Michael. Now that the quarter is behind us, it makes sense to pull back and look at the wider landscape we're working within. For a number of years now, the investment community has fixated on how quickly AI infrastructure, data centers, GPUs and large language models were being built out; that digital layer is scaling fast today, and rivalry throughout the ecosystem keeps heating up. Meanwhile, all across Wall Street, a new sentiment is emerging, genuine anxiety about disruption. AI serves as a growth engine, yes, but it also acts as a deflationary pressure. Software and services categories in their entirety now confront doubt regarding pricing power, competitive moats and, in the end, terminal value; as AI accelerates everything, plenty of business models are getting repriced and, for some, structurally impaired. Looking at history, foundational technologies tend to generate value in layers. Electricity's arrival didn't just supply power to what already existed. It restructured the physical economy and made room for brand-new industries. AI will probably trace a comparable trajectory. The coming stage of AI is physical. Intelligence is shifting out of the data centers toward the edge, into robotics, advanced manufacturing, autonomous systems, defense platforms and electrified mobility. Once intelligence takes embodied form, actuation and motion become necessary. Motion calls for magnets. Rare earth magnetics aren't some application layer that algorithmic substitution can threaten. They're physical infrastructure, indispensable inputs that turn electrical energy into movement that is precise and efficient. As intelligence extends into the physical economy, magnet intensity per system goes up rather than down, since precision, torque density and energy efficiency grow more crucial as systems gain autonomy. Against this backdrop, our positioning differs fundamentally from that of a large portion of the market. Plenty of companies are wrestling with doubts about their long-term economic durability in an AI world, while we are building straight into its structural expansion. What we hold isn't code that can be swapped out overnight. It's scarce, strategic industrial capacity whose economic and national importance keeps rising, capacity that gains value as the physical AI cycle matures. At moments such as this, leadership matters, the nations and businesses controlling essential industrial capacity determine the direction of this new era. We built MP to serve as precisely that sort of platform, vertically integrated, technically differentiated, scaled and American. Discipline and a focus on execution will remain our priorities. Eventually, structural demand combined with thoughtful capital allocation is what produces enduring value; that is the work we're undertaking deliberately, methodically and with a long-term horizon; with that, I'll hand it back to the team for Q&A.
That was really useful, and congrats on the first quarter of getting significantly better pricing under your PPA arrangement. If I may, I'd like to ask one question plus a follow-up. So regarding the OEM, could you say whether it's an American auto OEM or one based overseas?
This is Ryan. As Jim said in his prepared remarks, we won't be going into specifics, but this is among America's top technology companies. Broadly, what this deal shows is something we've discussed for quite some time now — the growing need for NdPr, and the worth of the platform we offer, since we can serve as a solutions provider for these kinds of businesses that want to move their supply chain away from China. Being able to handle the different stages of the supply chain, from raw materials right through to magnets, is, I think, extremely important.
All right, that's reasonable. I get why you'd want to hold back on specifics. Now, regarding the schedule for 10X — it feels like everything is progressing at quite a pace. Initially, 2029 was viewed as something like the goal year. So could that be pulled earlier? Is it conceivable that 10X might end up being built a bit faster than your original expectations?
So, with 10X, the sourcing, the emphasis on long-lead equipment, plus a great deal of design and engineering work — all of that genuinely got underway, Lawson, in July of last year. Right after the agreement was signed, we began assembling the new team in that location, getting that new team aligned with the team we already had, and then considering every piece of equipment and every part of the process we could tackle while we proceeded with our site selection process. What I can tell you is that I've instructed the team to approach this project — you may have come across the phrase 0-based budgeting; well, this is a 0-based days project. Our whole focus is on bringing this online as fast as we can. We've stated that commissioning will happen in 2028. Considerable progress has been made. We believe it's on track, and we'll simply — we'll continue attempting to recover as much time as possible, as forcefully as we can, and things are going very well.
I'd like to revisit that OEM agreement. When you consider the P&L and the economics of selling oxide straight versus selling Magnets, how do you weigh the trade-off between those 2?
George, this is Ryan. What matters most, I believe, is that the way we're handling this is that Independence, for all practical purposes, is fully committed at this point, and I believe we've made our approach regarding 10X quite clear. Because of the offtake already in place there, we enjoy the advantage of staying very deliberate about how we bring commercial syndication out of that facility. As I said before, what's key is that among all these different OEMs — the ones truly speeding up their examination of their supply chain and trying to move away from China as fast as they can — we are genuinely the sole solutions provider. So the ability to be a big player there, supplying raw materials into their supply chain right away and capturing that value now, while also unlocking some fairly substantial possibilities downstream, I'd anticipate, strikes me as enormously value accretive for us. That's why, as I noted during my prepared remarks, securing an oxide agreement at this magnitude definitely benefits the Materials segment in the near term — from a working capital standpoint, from the timing of sales, and naturally from the value we'll capture there through our product, given the different stages of development we're at as we scale Mountain Pass toward its eventual full run rate, even as we keep ramping Independence carefully and bring 10X online.
Thanks. For a follow-up, I would really like Jim's take on what he believes is going on with NdPr pricing in China.
George, as you're aware, commodity pricing is never easy. There are, I believe, a handful of factors here. First, in the wake of our deal, I believe the Chinese will not act so mercantilistically anymore, since there's nothing to gain. And finally, I'd add — I'd also add that NdPr is experiencing demand based on what we observe regarding physical AI and, as I noted during the prepared remarks, the considerable enthusiasm surrounding everything taking place with motion and the economy; and therefore, I believe — when you consider critical minerals, there's a form of scarcity. You have 2 varieties. There are the heavies and certain other specialty metals that are simply scarce, and we must source more of them here since China is the sole choice; yet broadly speaking, the demand curve for those tends to be fairly steady or flat; on the other hand, with NdPr, particularly given the limits on heavies, what we observe — and we're witnessing it in real time. I suspect we could offer a few current examples in which we've seen a significant rare earth magnet consumer alter their specification overnight, basically to remove heavies and concentrate on moving toward greater NdPr. So what we observe is the traction motor, the EV and hybrid manufacturers concentrating on attempting to obtain — to maximize NdPr, moving away from heavies. And in fact, we're seeing, once more, immediate action on that front that matters for that, plus all the remaining applications — I believe NdPr will clearly be the essential material for physical AI, and that's the significant growth driver in that area. So when you total everything up, I believe, first, it's the market's response to achieving a somewhat more normalized NdPr price, since we're moving nearer to market economics. We haven't reached that point; and second, it's a result of certain current applications, in which we observe real-time substitution toward greater NdPr. And third, it's much of the growth enthusiasm we're witnessing across the board. And I'd note, I brought this up — I alluded to this during the prepared remarks, but a sort of bifurcation is occurring. I believe NdPr is truly the growth commodity, and I think we — once more, commodity prices can always behave wildly, but I believe we will witness ongoing acceleration in NdPr prices. And I'd add that the market environment right now perhaps places too much weight on pricing for the heavies. I think there will ultimately be multiple sources of heavies worldwide; we're definitely devoting quite a bit of effort to that matter. As Michael said, our heavy separation will be online this year. So my guess is that within a handful of years, we could easily see that market become pretty well saturated, and that wouldn't shock me at all.
I suppose this is somewhat of a follow-up to what George asked. In the market, there's a great deal of attention — and I think policy is driving this too — on things like price floors. But Jim, when you gave your prepared remarks, you made it sound as though there's upside in NdPr that deserves greater recognition and more attention, and we've indeed witnessed some firmness lately. Is there perhaps a way to frame what that NdPr price level or entitlement might look like? You mentioned something like a normalized price. What might that actually be? Or what might that range look like across time? And is there some trigger or timeline by which you believe that could finally come to pass?
That's really such a difficult matter. There's that well-known saying that forecasts reveal more about the forecaster than about what lies ahead, but regarding NdPr, I believe I've consistently put forward the position that if we were truly operating in what you might call a genuine fair — free market for NdPr, prices would be substantially higher, far above where they stand now, certainly reaching into the hundreds of dollars, purely due to the — what constitutes a sufficient return on capital needed to get projects up and running, correct? If we — if no nonmarket geopolitical player were involved in price setting and it were simply a question of what would draw capital in, I believe prices would be substantially higher. Yet the market's realities are the market's realities. We don't have a — we've achieved considerable progress, clearly, considering what we've accomplished with DOW and considering certain market developments, but plainly we aren't entirely there yet. And then, on the following point, it really boils down to the pace at which physical AI matures. I think we — clearly, you saw today's announcement of a massive strategic partnership concerning offtake. I've actually been, I'd say, genuinely and pleasantly taken aback by how thrilling the level of engagement has been with various companies, a great many of which sit in the physical AI space or aim to be working behind the scenes. So I do believe that will genuinely generate yet another leg higher, though it's difficult to gauge how rapidly that will come online. But I'd urge you, you can clearly run the numbers yourself, build your own curves regarding what that means for the market, and then obviously, from a supply chain perspective, what will unfold in the West as opposed to what constitutes Chinese supply for China, especially since I think robotics is genuinely a dual-use technology, meaning it will be genuinely difficult for many Western companies to secure licenses for rare earth magnets. And we are witnessing this play out right now: there is genuine anxiety among the largest and strongest tech firms globally—those eager to compete—that licenses for this won't come through. Clearly, that puts us in a very strong position.
Yeah, that makes a ton of sense. And I just want to follow up on a few of those policy items that are still floating around the sector, like Project Vault and Section 232, among other things. Do any of those have implications for you guys? Is it affecting the conversations you're having with non-DOW offtake counterparties, or perhaps even the discussions with DOW? And would you anticipate any effect on pricing? I realize it's early and there's a lot of uncertainty. But maybe you have some thoughts on how you could potentially stand to gain from that?
Yeah. No, that's a good point. I'd say first and foremost, clearly, the transaction we did last year set off a major shift in our sector, and it unlocked the capital markets. And obviously it gave us a wider opening here in the West to tackle this challenge, and that's exciting. But I think what you're witnessing is that it isn't only DOW — it's every part of the administration, the key verticals, be it state, commerce, treasury, interior, or energy. This topic has everyone's attention; it's a central directive coming from the President. So I think as you look across a number of these initiatives, you brought up Vault, and there's FORGE, and there are several others, I'd characterize them as being at differing stages of maturity, each one. A few of them need allied countries to essentially reach a position and establish a plan. Others involve public-private coordination. What I'd tell you is that most of these efforts are, I'd say, relatively early or at minimum not finished. We'd anticipate playing a significant role in how some of those areas you referenced take shape. So my sense is that all of them would be quite favorable for us, but there isn't anything concrete at this point that I could point you toward specifically.
Can you walk us through how you expect CapEx to be spread over the year — you've guided to $500 million to $600 million, with most of it tied to 10X — and how that might look quarter by quarter? Same kind of question on the EBITDA cadence.
Yeah, absolutely, Corinne. It's Ryan here. As we think about the cadence, it's going to be fairly lumpy throughout the year within that range, because we'll probably see some capital spending up front, particularly for the land we just acquired — that was announced this morning — for our Northlake, Texas site. As the year progresses, obviously, the portion of that CapEx directed at 10X will start to ramp as construction gets going. We also have other parts of the company and growth efforts we're working on — in particular, what we're doing with Apple — and those call for capital spending basically across the year, though it probably tilts a bit toward the later months, once that equipment is being put in place and readied for the 2027 start of that agreement. On the EBITDA cadence, clearly, to return to what Jim said about forecasting NdPr prices, my sense is that's probably the main factor at play. As you know, we hold onto all of the benefit from higher NdPr pricing until 10X is commissioned. So should prices keep doing what they've been doing and keep climbing, then earnings would move in line with that, plainly. Like I mentioned earlier, the way our contracts work is that pricing generally reflects a lag of roughly one quarter, sometimes a bit more, which means the effect shows up gradually rather than right away. The spot price you're seeing today, then, won't be reflected in Q1 — it'll show up in Q2 instead. That's the key thing to remember when you're thinking about how the quarters line up.
And for my second question, if I could. Could you talk about what you see as perhaps the ideal route to reach those substantial requirements past what you currently possess? For instance, are you considering additional recycling? Are you considering expanding the mine, and are you evaluating every peer, M&A? I'm just attempting to determine.
I believe, apologies, Corinne, is what you're asking how we intend to increase supply? Or you mentioned additional recycling. And what else did you say?
Yeah, so particularly on the heavy side, how would you go about that? Would that be via M&A, or something like recycling — recycled content?
Thanks, Corinne. Michael here. We're definitely considering recycling as one route, and as I've mentioned before, I believe the integrated setup at our site lets us tap into a wide variety of feedstocks. Through our deal with the Department of War, they back us in sourcing feedstocks globally; so I'd expect a blend of maybe conventional recycling plus nontraditional feedstocks. But like Jim said, the headway we've made in cutting our heavy requirement means we're less worried on the whole, yet we expect to be the biggest heavy rare earths producer in the Western Hemisphere, or from now on for as far ahead as you can picture.
Congratulations on how far the various projects have come. Regarding 10X, I just want a clarification about the CapEx. Your announcement mentioned $1.25 billion plus. Could you give us a bit of color on what that "plus" means? Help us size it, quantify it, for modeling purposes. Then there was also something in that release indicating that 100% of the light and heavy rare earth needs for 10X and Independence would be supplied by the company's Mountain Pass operation. That would mean no real reliance on third-party heavy rare earths. Given how important that point is, I just wanted to confirm it.
Yes, of course, Carlos. Let me take your second question first. For both facilities, we do plan to cover all heavy and light feedstock from the Mountain Pass processing facility. As Michael just described, we have a number of initiatives underway alongside the department of War, alongside Apple, plus our own sourcing strategies, to secure additional heavy rare earth feedstock. Another thing that matters greatly here is this: if your business model leans on third-party feedstocks to supplement what you can produce, then producing both lights and heavies is essential to being competitive in that market. Competing economically is impossible unless you can capture value from every one of those. So we have a great deal of confidence in our continued ability to source third-party feedstock, which, honestly, we are doing right now. Turning to your question about 10X and the capital investment, because of how significant this is and its scale, our focus is on doing what is right for our core customer, for our future customers, and from a return on invested capital standpoint. The reality, then, is that certain portions of that facility might expand or contract as time goes on, once we reach our final ultimate design. What we have consistently done, and will keep doing, is give all of you guidance annually so you can track our progress there. As we discussed, this year falls in the $500 million to $600 million range, which clearly represents a fairly substantial initial investment to bring that facility online as fast as possible.
Two fast items. To start, could you share where things stand regarding the Saudi JV, and how much capacity do you have to pursue another front JV, perhaps in Europe or South America? Then second, could you update us on how you're going about bringing formulation engineers on board, and what your plan is regarding the quantity of SKUs offered to customers? Or do you push them toward concentrating on a limited set of SKUs?
Absolutely. Regarding Maaden, there are essentially two paths. As you're aware, a binding term sheet was executed by us late last year. We need to complete the formal paperwork with our counterparty in that deal, the Department of War, and afterward with Maaden, in order to form the combined JV. Naturally, many of the remaining specifics must tie back to our finalized estimate for the build's cost, plus the facility's process flow. Quite a few elements remain for us to address before that side reaches completion. And naturally, this means the other work stream involves, from the operations side, Michael and his group handling it — the process flow, every element of how that facility will operate. So there are basically two primary streams in play. What I'd say is that this will clearly be a major, thrilling project, yet it will require time. I mean, we intend to ensure it gets built properly, and we'll proceed carefully and deliberately. So we'll go about it in a disciplined manner, I suppose that's the clearest way to put it, and we keep working on it. I believe the following question concerned South America or Europe. I mean, as you're aware, as an organization, we're opportunistic. Consider 5 years back from today when — approximately when we became public, when we were purely a concentrate business, we hadn't a single employee in our Magnetics business. Then jump ahead to now, and consider the size and depth of our organization and everything we've constructed. So I believe we can handle plenty of challenges, especially since I think we're heavily execution oriented, and we — we act swiftly and forcefully. So we'll stay opportunistic and I think, ideally, we'd never end up biting off more than we can chew. So I suppose that's a lengthy way of stating, certainly, we'll remain opportunistic about plenty of things. We definitely have plenty on our plate at present. I think you also mentioned recruiting and engineers, and we've definitely assembled a fairly remarkable team within our Magnetics business. That team keeps expanding. Completing that project is hardly a small undertaking. I'm clearly very focused on it. Since inking that agreement back in July, we've assembled an entire group for it, and that effort is now actively moving forward and progressing quite nicely. So our attention is squarely on getting it done.
The one point I would make in response to that, Laurence, regarding your question about concentrating on SKUs, is this — what strikes me as interesting is that, when you look at it from the standpoint of magnet formulation, Jim raised this earlier, there has been an enormous push from our customers toward really tightening their attention on formulations that are heavy rare earth-free or have sharply reduced heavy rare earth content. Jim likewise discussed the headway we've achieved on that front, even measured against our original formula for EV traction motors, which as an illustration sits 60% under where we began, and that starting point was already toward the lower end of what you'd call typical market levels. What we're also in the middle of working through at present, I'd say, is that basically, within the Magnetics business, the sole path to making the economics function during the early phase is possessing scale and being capable of establishing the quality and production quality processes that a customer such as an automaker requires. That is the thing that lets you handle everything else and bring in the added industrial complexity of the many different SKUs. The wonderful part, I'd say, is that with the foundation beneath us from these enormous customer agreements with GM and Apple, for instance, 10X is built to carry a substantial degree of added flexibility. Should you have looked at the agreements we filed with the Department of War, they referenced a range of magnet specifications and grades that we'd mutually agree upon. We've already accomplished that alongside them, and we essentially unlocked, I'd say, the large bulk of the market in that area along with what sits in our development pipeline. So we are certainly developing the capability to handle differing formulations and differing SKUs across various sizes and shapes as we move toward 10X.
Nice work on the numbers and all the useful details you shared. I'd like to ask about offtake, and I suppose, about additional offtake deals. My guess is that you want to, and you have the bandwidth or the capacity, to back more offtake agreements. Or is that going to be constrained by growth, like Upstream 60K, or by getting the 10X facility operational? I think this also ties into what Jim mentioned before regarding some of your customers focusing clearly on physical AI as opposed to possibly auto, where volumes might be far bigger; so do you have any sense of how we should think about offtake agreements going forward?
All right, Ryan, why don't you kick things off with offtake, and then I will...
Absolutely. Yeah, Bill, our approach stays opportunity-driven, I'd say. When you look at how much demand the automakers have, roughly 30% of magnet content these days is used in the auto supply chain, and close to 100% of it right now originates in China. So the portion of that pie up for grabs, as those customers try to diversify where they source from in the current market, is really remarkable. To echo what Jim said, we're beginning to witness a much more rapid focus from many other kinds of users. I mean, all sorts of physical AI applications, be it robotics and those sorts of uses, or honestly, plenty of the components that end up in data centers are big magnet consumers too; so there are many growth avenues in this segment at the moment that we're also aiming at, but we'll stay opportunistic. I'm not sure whether Jim or Mike would like to add anything further.
One more point I'd make: our upstream operations should be getting close to 6,000 tons per annum by the time this year ends, and against that, the demand from our magnet side alone means there's still space for further offtake demand. On top of that, as we scale up third-party feedstocks along with recycling, that gives us extra material we can keep supplying to the market.
Then for my follow-up, could you give us an update regarding the magnet qualification at your lead auto customer, plus the remarks you made, Michael, about the troubleshoot ramp? I'd like to dig into that further. Is there anything worth highlighting when it comes to a focus on yields or throughputs? Or I believe the technical performance is okay; but is there anything else you can address regarding that — regarding the magnet ramp?
Let me begin with the second part. Ryan will take the PPAP; when it comes to the magnet ramp, taking a step back and considering where we stood several years ago, I couldn't feel more proud, and I also couldn't feel more optimistic that our team is going to be able to take on this challenge. We understand that launching a new facility is difficult, but the challenges we've achieved — gotten past thus far compared with what remains for us to do now, I think it's simply a matter of focus on execution. And that's what we've carried out in Mountain Pass, focus on quality and focus on execution for the customer and over time, build the volume. So we'll keep running that playbook.
Absolutely. And regarding PPAP, Bill, like I mentioned before, stepping back a bit, one factor behind our choice to begin with an automotive client was that we aimed to make magnets from the start capable of meeting some of the toughest quality criteria needed for applications such as EV motors. Clearly, PPAP is a pretty thorough procedure. It examines not just product quality but the manufacturing processes too. You must show various run rate scenarios, and so on. So honestly, as we go through that, we'll keep — when we open a facility, we'll refine our processes and abilities, and that will position us to speed up and onboard customers such as Apple even faster. So we're only at the start of that process. Yet because of our considerable involvement from the very beginning in constructing this plant alongside GM, we think it will be a somewhat quicker process than what you'd see with a standard auto supplier qualification. And that's what shapes our guidance for the latter half of the year concerning salable magnets.
Thank you. Well, thanks to all of you. It was a pretty remarkable quarter, capping off a truly transformative 2025, and 2026 has begun very strongly; so we'll return to our work and talk with you again before long.
Read the full transcript via the API
You're viewing the first half of this call. Get the complete MP Materials Corp. transcript - plus 132,000+ transcripts from 7,000+ companies, speaker segments and full-text search - through the EarningsAPI REST API or hosted MCP server.
Get an API key View API docs →For developers and AI pipelines
Programmatic access to MP Materials Corp. earnings transcripts and 132,000+ others is available through the
EarningsAPI REST API and the hosted MCP server.
Quarterly plans from $162 - full transcripts, speaker segments, full-text search,
and the /api/v1/transcripts/recent polling endpoint for ETL pipelines.