MP Materials — Pass-with-trigger — 2026-06-09

Refreshed 2026-06-10 — EPV floor recomputed from the corrected consolidated adjusted EBITDA (audit-log #013); central value, trigger, and verdict unchanged. Prior version: 2026-06-10-epv-recompute.

1. Bottom line

MP Materials owns the only scaled rare-earth mine and processing complex in the United States, and the Department of Defense sits underneath it as the largest shareholder with a $110/kg price floor. That makes the downside unusually hard for a cyclical. It does not make the stock cheap. At $58 the market pays roughly ten times the earnings power in the ground today and books the rest as credit for a magnet ramp that does not produce commercial output until 2028. The honest central value, anchored to assets and the price floor rather than to a magnet business that does not yet earn, is about $50 — below the current price. The variant perception is that this is Phase 2 of a capital cycle, not a permanent regime change, and the $58 quote prices the cycle as if Phase 3 will not arrive. The verdict is pass-with-trigger at $42, the level where a buyer pays for the asset and net cash and gets the magnet optionality for free. No position today.

2. The business

What they do

MP Materials mines and processes rare-earth elements at Mountain Pass, California — the only operating rare-earth mine in the United States and one of the few scaled processing sites outside China T1. It sells separated rare-earth oxides, chiefly neodymium-praseodymium (NdPr) oxide, the input to the permanent magnets that drive electric-vehicle motors, wind turbines, robotics, and defense systems. The company is integrating forward into magnet manufacturing at its Independence facility in Fort Worth, Texas, and is building a much larger plant, internally called 10X, with first commercial product expected in 2028 T2.

How they make money

Two segments. Materials sells separated oxides and concentrate; it carries the mine's high fixed-cost, capital-intensive economics, so unit profitability swings hard with the NdPr price. Magnetics is the early-stage magnet business, which booked its first revenue this year. In the March 2026 quarter the company's two segments generated $36.7 million of Materials adjusted EBITDA and $9.6 million of Magnetics adjusted EBITDA; consolidated adjusted EBITDA was $36.6 million after corporate/unallocated costs [corrected 2026-06-10 — see audit-log #013: the original text below stated "$46.3 million of adjusted EBITDA," which is the sum of the two segment figures, not the company's consolidated adjusted EBITDA. MP's actual consolidated adjusted EBITDA was $36.6M per the Q1 2026 8-K; the vault's own 2026-06-09-PM research note correctly recorded "+$36.6M adjusted EBITDA." The error overstates the EPV/EV-EBITDA inputs below — see the markers in §5 and the Tier 2 Backlog item] the company earned $46.3 million of adjusted EBITDA — $36.7 million from Materials and $9.6 million from Magnetics T1. The economic engine today is the oxide, sold into a price that has nearly doubled in five months; the magnet business is a forward bet funded by outside capital, not a current profit source.

Segments and geographic mix

Revenue is concentrated at Mountain Pass and increasingly at Independence. The Department of Defense set a $110/kg floor on NdPr oxide MP produces, and agreed that for ten years after the 10X facility is built, 100% of its magnet output is purchased by defense and commercial customers with shared upside T2. Apple separately committed $500 million for magnets made from recycled feedstock at Independence, with shipments starting in 2027 T1.

Recent history

The March 2026 quarter was a record. Revenue reached $132.9 million, up about 49% year over year, on NdPr oxide output of 917 metric tons, up 63% year over year T1. Two forces drove it. Volume rose as Stage II processing matured, and price rose as NdPr oxide climbed from roughly $53/kg in January 2026 to near $100/kg by May, after China's April 2025 controls on heavy rare earths and Beijing's May 2026 reaffirmation that those controls are here to stay T3. The company is spending more than its operating cash flow on the magnet ramp, funded by the DoD package, a $150 million strategic-capital loan, and a $1 billion J.P. Morgan / Goldman Sachs facility T2.

3. Structural assessment

Moat

• Type: A supply-side scale and government-backstop advantage — the only US scaled rare-earth-to-magnet asset, paired with a sovereign customer who has set a price floor and an offtake commitment.
• Mechanism: Mountain Pass cannot be replicated quickly. A new mine plus separation plus magnet line runs four to six years from a standing start and billions in capital T2. The DoD relationship is harder still to copy — a competitor can build a plant but cannot manufacture a Pentagon equity stake and a price floor.
• Evidence in numbers: The moat is visible in the price floor ($110/kg, above current spot near $100) and in the capital others must spend to approach the same position — USA Rare Earth raised roughly $3.1 billion to build a magnet footprint smaller than MP's planned capacity T1. It is not yet visible in return on invested capital, because the magnet business does not yet earn.
• Power and width: The return-over-cost-of-capital spread on current operations is thin once Mountain Pass depreciation is charged honestly. The width is contested: the same Phase 2 capital influx that confirms the theme is funding the competitors who erode the scarcity.
• Decay risks: Chinese controls relaxing, Western capacity arriving on schedule, or magnet-free motor designs all narrow the moat. The 2010 cycle is the cautionary base rate — Mountain Pass's prior owner, Molycorp, went bankrupt in 2015 after the last rare-earth price spike collapsed.

Bottlenecks

MP sits at the chokepoint of the Western magnet supply chain — separated NdPr oxide and, increasingly, finished magnets. Upstream it depends on its own ore body, which removes feedstock risk. The bottleneck it owns is separation and magnet-making capacity outside China, which is genuinely scarce today. Downstream it is selling into defense primes and Apple under long contracts, which secures volume but caps the rent it can extract, because those buyers negotiated shared upside, not open-market pricing.

Industry structure

China controls roughly 60–70% of rare-earth mining and over 85% of processing T3. That concentration is the whole investment case and its whole risk. Western capacity is in the steep part of its build — billion-dollar equity rounds, sovereign stakes, bank lines, government loans, and forced divestments of Chinese-linked holders. On the Chancellor capital-cycle framework this is textbook Phase 2: returns above cost of capital pull in capital, capacity follows, and returns compress when it arrives T2. The cycle in light rare earths historically runs seven to ten years; the 10X first product (2028) and the cohort's other plants (2026–2028) mark when supply pressure starts to relax.

4. Variant perception

What consensus believes

The sell side rates MP a near-unanimous Strong Buy with a mean target around $79 and a range from roughly $62 to $100 T3. The implicit case is that Chinese controls are durable, the magnet ramp executes on the announced timeline, MP captures a disproportionate share of a structurally short market, and the price floor is a backstop under a price that will mostly sit well above it.

What you believe differently

The $58 price embeds a stack of assumptions whose joint probability is lower than the cohort multiple implies. Specifically: that NdPr does not compress as Western capacity arrives through 2027–2028, that the 10X ramp hits its 2028 milestone on cost and schedule, and that the November 2026 expiry of the suspended Chinese control package resolves in favor of continued tightness. This is a structural and forecast variant. The deep-value frame credits the oxide earnings power and the asset, treats the magnet ramp as optionality rather than as base-case value, and concludes that the central value is about $50 — meaningfully below where the cohort trades.

Steelman of consensus

The strongest single argument for consensus is that the demand picture is qualitatively larger than 2010 and the supply position in heavy rare earths is genuinely absent outside China. Adamas Intelligence frames a roughly 110,000-tonne 2030 structural magnet deficit against only 11 of 42 needed facilities in development T2. If that deficit is real and durable, the surviving Western producers earn rent for a decade, and MP, with the floor and the offtakes, is the surviving producer. The reason to reject it as a base case is not that it is wrong but that it is a bull tail being priced as the expected outcome, against a base rate that says the last comparable spike collapsed in four to five years.

Gap-closer

Two events resolve the variant inside twelve months. The Q2 2026 print (early August) tests whether NdPr realizations and magnet revenue track the ramp. The November 2026 expiry of the broader suspended Chinese control package is the cleanest binary catalyst in either direction.

5. Valuation triangulation

Method 1 — EPV (no growth)

Recomputed 2026-06-10 from the corrected consolidated figure per audit-log #013; derivation and prior state in 2026-06-10-epv-recompute.

Consolidated adjusted EBITDA for the March 2026 quarter was $36.6 million; annualized, roughly $146 million T1. The GAAP cross-check per MP-fundamentals: GAAP-derived EBITDA for the same quarter was $8.0 million, and the gap is the add-backs — $12.9 million of stock compensation plus start-up and other adjustments — that the adjusted figure credits and the GAAP figure does not T1(/brain/mp-fundamentals)]. Mountain Pass depreciation is heavy. Charging an honest $90–110 million against the $146 million leaves normalized EBIT of roughly $36–56 million; for reference, FY2025 depreciation was $89.3 million and the Q1 2026 run-rate is about $128 million annualized as buildout assets enter service T1(/brain/mp-fundamentals)]. Taxed at 21% and capitalized at an 8–10% cost of capital, the operating business is worth roughly $0.3–0.55 billion. Add net cash of about $0.7 billion — carried from the original thesis against the 8-K, now under the balance-sheet flag below — and divide by 178.0 million diluted shares, and EPV lands near $5.50–7 per share, central about $6 T1(/brain/mp-fundamentals); AS-cal: normalization and WACC bands]. This band replaces the three figures previously in circulation — the $8–12 published here, the shadow matrix's $11, and the builder note's $7.50 — all of which descended from the overstated $185 million input. The floor sits even further below the price than first written. The $110/kg DoD floor lifts the durability of the number, not the level.

Balance-sheet flag. The pipeline cannot yet confirm the net-cash figure. Per MP-fundamentals, cash was $886.3 million against total debt of $1.00 billion at quarter end — net debt of $114.2 million on a cash-only read — but the concept map carries no short-term-investments tag for MP, so investments held outside cash are invisible to it. The ~$0.7 billion stands on the 8-K until the concept map is extended and the fundamentals file regenerated (queued in the Backlog). Sensitivity: each $100 million of net liquidity that fails to verify removes about $0.55 from the band; at the cash-only read the band bottoms near $1–2.50. No outcome in that range moves the floor toward $58 or touches the asset-anchored trigger.

Method 2 — Asset / replacement value

Total shareholder equity is about $2.4 billion, or roughly $13.5 per share of book T1. Book understates the asset. The integrated US rare-earth-to-magnet platform is being reproduced elsewhere at billions of dollars and four-to-six-year lead times; a strategic buyer would not replace Mountain Pass plus Stage II separation plus the Independence magnet line for book value. A replacement / private-market estimate, including the strategic scarcity premium and the DoD relationship, supports roughly $30–42 per share. Net cash of about $4 per share sits on top. This is the load-bearing floor for the trigger — below roughly $42 the buyer pays for the asset and gets the magnet ramp for free.

Method 3 — DCF / reverse-DCF (cross-check)

A forward DCF that credits the 10X ramp, the Apple and DoD offtakes, and NdPr prices holding near the floor produces a value in the $55–85 range, which is where the sell side sits. Run in reverse, the $58 price implies the 10X facility ramps on schedule, the price floor is a floor under a higher market price rather than the operative price, magnet revenue scales as guided through 2027–2028, and there is no Phase 3 compression in NdPr through at least 2030 T2. That is a defensible bull case but not a margin-of-safety entry.

Method 4 — Comparables

At $58 the enterprise value is about $9.5 billion against roughly $146 million of annualized consolidated adjusted EBITDA, an EV/EBITDA multiple near 65x T1. Lynas, the closest scaled peer, trades on prices and volumes that are also cycle-elevated. The cohort multiple is pricing 2028 capacity, not current earnings, which is the definition of a Phase 2 multiple.

Method 5 — PMV / replacement value

Covered in Method 2. The strategic-buyer anchor is the most relevant cross-check for this name because the going-concern earnings frame fails — there is no other US asset a defense-aligned buyer could acquire to replicate the position.

Triangulation

The methods disagree by design. EPV says $5.50–7, asset/replacement says $30–42, the optionality-credited DCF says $55–85. The deep-value central weights the asset and the floor heavily and credits the magnet optionality conservatively, landing at a central value of $50, with a range of $35 (bear: Phase 3 compression with the floor and net cash holding) to $95 (bull: controls extend, ramp executes, offtakes compound). The range is wide because almost all of the value above the EPV floor is optionality, and the dispersion reflects honest low confidence.

Margin of safety

The name is cyclical and contested, which normally demands a 40–50% discount to central value. The hard DoD floor plus net cash justify a tighter discount than pure cyclical historicals — the downside is genuinely backstopped in a way Molycorp's never was. At a central value of $50, the re-engagement trigger is $42, a roughly 16% discount that coincides with the asset-and-floor backstop. At today's $58 the price sits about 16% above central value and well above the trigger. There is no margin of safety.

Shadow valuation matrix

Mandatory section. See shadow-valuation-matrix for methodology; full detail in MP-shadow-matrix.

• Spread: $6 to $85 (ratio ~14x).
• At-current-price verdict spread: the four split. Klarman passes, Greenwald and Buffett-modern wait, Mauboussin buys. Only the most growth-generous frame acts at $58.
• Where the spread comes from: almost entirely the growth-value credit and the runway band. Every dollar above the ~$6 EPV floor is the magnet ramp, and the methodologies differ precisely on how many years of above-cost-of-capital reinvestment to credit — the variable the 2010 base rate makes most contestable.

6. Consensus & gap

Gap to consensus mean exceeds 25%; full treatment in MP-consensus-gap.

Consensus map

The sell-side mean sits about 58% above AlphaSteve. Simply Wall St's intrinsic estimate sits essentially on top of ours, which is the tell — the gap to the street is a horizon and methodology gap (12-month momentum-credited targets versus an intrinsic asset-anchored frame), not evidence that one side has data the other lacks.

Classification of the gap

Structural — methodology choice. The gap is the EPV-plus-asset frame declining to credit optionality that the sell-side DCFs and 12-month targets credit through the multiple. The decomposition lives in MP-consensus-gap.

7. Quality and management

Capital allocation

Management has converted a single distressed asset into a vertically integrating platform and pulled in the Pentagon, Apple, and two bulge-bracket lenders as funding partners on terms that protect the downside T2. The reinvestment is aggressive — spending above operating cash flow — but it is funded by outside capital with a price floor attached, which is a defensible way to finance a long-dated buildout. The open question is return on that capital, which is unproven until the magnet business earns.

Incentive alignment

The DoD's roughly 15% effective stake through a $400 million convertible preferred plus warrants aligns the largest shareholder with the survival of the asset and the persistence of the floor T3. That is an unusual and favorable alignment for a cyclical commodity producer.

Communication quality

Disclosure is specific on volumes, prices, and segment EBITDA, and the company files cleanly on EDGAR. No flags in the language.

Governance flags

The convertible preferred and warrant package creates future dilution that a buyer must model; it is disclosed, not hidden. No material governance red flags.

8. Macro and cycle context

Rate environment

A long-duration, capital-intensive buildout is sensitive to the cost of capital. The 2026 tape carries Fed-hike fears, which raises the discount rate on exactly the 2028 cash flows the price is crediting 2026-06-09-AM. Higher-for-longer is a headwind to the optionality the bulls are paying for.

Cycle position

Phase 2 of a capital cycle, as developed in 2026-05-29-critical-minerals-capital-cycle-dossier-v1. Returns are accelerating because capacity has not yet arrived; the framework's discipline says returns compress when it does, in 2027–2028 on the announced timelines.

Regulatory backdrop

The whole thesis turns on Chinese export policy. The April 2025 heavy-rare-earth controls are active; Beijing reaffirmed them in May 2026; the broader October 2025 package is suspended until November 2026 T3. That expiry is the cleanest forward catalyst in either direction.

9. Risk and pre-mortem

Top three bear scenarios

1. Phase 3 compression (probability ~35%). Western capacity arrives roughly on schedule and China relaxes controls under negotiation or WTO pressure; NdPr compresses toward and then through the floor. The floor protects MP's realized oxide price, but the equity multiple compresses hard. Plausible downside to $35.
2. Magnet ramp slips (probability ~20%). The 10X facility runs late or over budget, Apple and DoD offtake volumes ramp slower than guided, and the optionality the price embeds is pushed out. Downside to $40 as the market re-rates the timeline.
3. Demand substitution (probability ~15%). Magnet-free motor designs gain traction and the 2030 deficit narrative softens. Slower-moving but structural; downside to the asset floor.

Permanent loss scenarios

A true zero is remote given net cash, the asset, the floor, and a sovereign shareholder. The realistic permanent-impairment path is buying the Phase 2 premium at $58 and watching multiple compression take the equity to the asset floor while the thesis stays nominally intact — the Molycorp pattern, but with a survivable balance sheet.

Kill criteria

• China formally lifts the April 2025 heavy-rare-earth controls before or at the November 2026 expiry.
• NdPr oxide spot falls and holds below the $110/kg floor, signaling the floor is now the operative price rather than a backstop.
• The 10X facility milestone slips materially past 2028 on company disclosure.
• Net cash turns negative without a corresponding step-up in magnet revenue.

Tail and macro

A sharp risk-off — a memory-led or rate-led equity drawdown — could deliver the trigger intraday without any change in fundamentals, which is the constructive version of the tail. The destructive version is a US-China rapprochement that prices out the scarcity premium overnight.

10. Position sizing

No position today; verdict is pass-with-trigger. On a fire at $42, this is a probe-to-mid name, not a core position — the wide value range and the contested moat cap conviction. Initial size on trigger would be in the 1–2% range, built in two tranches ($42 and a second at the asset floor near $35 if it prints), with a hard single-name cap well inside portfolio limits. The DoD floor and net cash are what justify acting at all on a cyclical the deep-value frame would otherwise pass entirely. Sell discipline on a future position: trim into central value, exit on a kill-criterion fire.

11. What you don't know

1. Whether China relaxes controls at the November 2026 expiry. Unknowable now; it is a policy decision. It is the single largest swing factor and resolves on a fixed date.
2. The realized return on the magnet capital. Not yet observable — the magnet business is pre-profit. The Q2 and Q3 2026 prints begin to show it. This is what separates the $50 central from the $85 bull case.
3. Where NdPr settles mid-cycle. The floor sets a contractual minimum but not the clearing price once Western capacity arrives. The 2010 base rate says lower than the spike; the demand picture says higher than 2015.
4. The dilution path from the DoD preferred and warrants. Disclosed but dependent on conversion timing and price.

12. The next steps

• Next review: Q2 2026 earnings, early August 2026 — test NdPr realizations and magnet revenue against the ramp.
• Calendar anchor: November 2026 expiry of the suspended Chinese control package — the binary catalyst.
• Watch for: any print of $42 (re-engagement) or $35 (second-tranche level); a kill-criterion fire; a material 10X schedule change.
• Watchlist action: carry MP at trigger $42, central $50, reviewed at the Q2 print or on a trigger approach within 10%.

Sources

• T1 MP Materials Q1 2026 8-K, SEC EDGAR — https://www.sec.gov/Archives/edgar/data/0001801368/000180136826000027/mpmcq12026er.htm
• T1 MP Materials 10-Q, period ending 2026-03-31, SEC EDGAR — https://www.sec.gov/Archives/edgar/data/0001801368/000180136826000029/mp-20260331.htm
• T1 MP Materials 8-K (Apple partnership), 2025-07-15, SEC EDGAR — https://www.sec.gov/Archives/edgar/data/0001801368/000119312525159093/d904727d8k.htm
• T1 MP Materials press release, "Transformational Public-Private Partnership with the Department of Defense," 2025-07-10 — https://mpmaterials.com/news/
• T1 USA Rare Earth Q1 2026 8-K, SEC EDGAR — https://www.sec.gov/Archives/edgar/data/0001970622/000121390026045339/ea028691001ex99-1.htm
• T2 Federation of American Scientists, "Unpacking the DoD and MP Materials Critical Minerals Partnership" — https://fas.org/publication/unpacking-dod-and-mp-partnership/
• T2 Adamas Intelligence, "Post-2030: Unfathomable Rare Earth Demand Growth Awaits" — https://www.adamasintel.com/unfathomable-rare-earth-demand-growth/
• T2 Chancellor (ed.), Capital Returns: Investing Through the Capital Cycle 2002–15, Palgrave Macmillan, 2015 — Part I
• T2 Mauboussin & Rappaport, Expectations Investing, HBSP 2001 — implied-expectations framing
• T3 CNBC, "Pentagon to become largest shareholder in rare earth magnet maker MP Materials," 2025-07-10 — https://www.cnbc.com/2025/07/10/pentagon-to-become-largest-shareholder-in-rare-earth-magnet-maker-mp-materials.html
• T3 TechCrunch, "Apple commits $500M to U.S.-based rare-earth recycling firm MP Materials," 2025-07-15 — https://techcrunch.com/2025/07/15/apple-commits-500m-to-u-s-based-rare-earth-recycling-firm-mp-materials/
• T3 GovCon Wire, "MP Materials, DOD Partner to Boost US Rare Earth Magnet Supply," 2025-07 — https://www.govconwire.com/articles/mp-materials-dod-public-private-partnership-rare-earth-magnet-10x-facility
• T3 StockTitan, "MP Materials Q1 2026 revenue jumps 49% with record NdPr output," 2026-05 — https://www.stocktitan.net/sec-filings/MP/
• T3 GuruFocus, "MP Materials Corp (MP) Q1 2026 Earnings Call Highlights," 2026-05 — https://www.gurufocus.com/news/8847298/
• T3 MarketBeat, MP analyst forecast, 2026-06 — https://www.marketbeat.com/stocks/NYSE/MP/forecast/
• T3 stockanalysis.com, MP forecast and statistics, 2026-06 — https://stockanalysis.com/stocks/mp/
• T3 rare-earth-mining.com NdPr price tracker, May 2026 — https://rare-earth-mining.com/neodymium-price/
• T3 US News, "China Says Rare Earth Controls Lawful," 2026-05-20 — https://www.usnews.com/news/top-news/articles/2026-05-20/
• T3 certivo.com, "China Rare Earth Export Controls 2026" — https://www.certivo.com/blog-details/china-rare-earth-export-controls-2026-what-new-licensing-rules-mean-for-manufacturers

Linked

• MP — landing file
• MP-fundamentals — EDGAR pipeline fundamentals (citation source for GAAP figures)
• 2026-06-10-epv-recompute — EPV recompute revision note
• MP-shadow-matrix
• MP-consensus-gap
• MP-calibration
• first-read-2026-06-08 — the seed first-read
• 2026-05-29-critical-minerals-capital-cycle-dossier-v1 — the theme dossier
• capital-cycle
• liquidation-and-asset-value
• earnings-power-value-greenwald
• margin-of-safety-pricing
• 2026-06-09-AM — source scan escalating the build