Stock Analysis · Bulls vs Bears

USA Rare Earth USAR

🐂 Bulls vs Bears 🐻Report #213 Oct 2026
★ Willem owns this stock · 1.5% of the portfolio
69
high risk▲ +4 vs last
🐂 9vs12 🐻
Bears lead
35%
▲ Chance of the biggest tailwind the forecasts miss
50%
▼ Chance of the biggest headwind the forecasts miss
The verdict

The evidence leans bearish for the next 6–12 months. USAR must raise about $675M by 2027-03-31 and a further $875M by 2027-12-31 under its Commerce agreement, with the stock at $11.82. Serra Verde produced only 40 t of rare-earth oxide in 2025, has no named non-China separation route for 2027, and cannot pay USAR dividends until it has repaid DFC debt. The long-term asset case is still real: a government-floored…

The Bulls vs Bears podcast is being made
Two AI hosts debate this report, NotebookLM-style. It appears here automatically within a few hours.

Written by Claude (Anthropic’s AI) from public sources · Willem owns USAR · AI can make mistakes — numbers in brackets point to the sources · Not financial advice

🐂 Bull arguments 9

  1. 1Floored heavy-rare-earth mine with 70% of the upsideStrong
  2. 2Western Dy and Tb trade at 4–6x Serra Verde's floorsMedium
  3. 3$1.39B pro-forma cash, net cash, no going-concern doubtStrong
  4. 4Washington owns equity and has $3B+ at stakeStrong
  5. 5DFARS 2027 ban meets a US market that is ~90% shortMedium
  6. 6Consensus already sits a third below management's planWeak
  7. 7Stock trades 45% below the PIPE and the Commerce warrantWeak
  8. 8Non-China midstream pieces: LCM metal, Carester, Wheat RidgeWeak
  9. 9A CEO who has actually run this mineWeak

🐻 Bear arguments 12

  1. 1~$1.55B must be raised by end-2027, starting ~$675M by MarchHigh
  2. 2Serra Verde is still ramping: 40 t produced in 2025High
  3. 3No named separation route for Serra Verde output after 2026High
  4. 4Serra Verde cash is locked behind DFC debt "for an extended period"Medium
  5. 5Not one dollar of the Commerce package has been disbursedMedium
  6. 6Every operating promise has slipped, under a third CEOHigh
  7. 7126M resale shares, 53M already free, 37M more on 12-02Medium
  8. 8Ex-China NdPr fell below $110/kg; light-rare-earth premium goneMedium
  9. 9Brazil's new law lets Brasília vet control and contractsMedium
  10. 10Rivals already hold binding magnet contracts and floorsMedium
  11. 11Round Top's grade is ~0.03% TREO; critics call it uneconomicMedium
  12. 12China blacklist, MP lawsuit and a Senate probe hang over itLow

Bull arguments show how well they survive the devil’s advocate (strong / medium / weak); bear arguments show the damage they could do after the counter-argument (high / medium / low).

Bull case

🐂 Bull #1Floored heavy-rare-earth mine with 70% of the upside

  • The point: Serra Verde's Phase-1 Nd, Pr, Dy and Tb output is sold to one US-government-backed buyer at escalating floor prices. Serra Verde keeps 70% of any market price above the floor. Downside is capped and the upside is mostly kept.
  • Evidence:
    • The buyer is the US SIIE, LLC SPV. It takes 100% of Phase-1 Nd, Pr, Dy and Tb for the earlier of 198M tonnes of ore or 20 years from commercial operations [4,8].
    • Floors are $110/kg NdPr, $575/kg Dy and $2,050/kg Tb, escalating 2% a year [10]. Serra Verde keeps 70% of the excess of market over floor, net of separation costs, and cost and yield variances are split 70/30 [4,16].
    • The SPV was capitalised at $1.55B on 2026-08-24, including $750M of US government funding (up from $500M) and a ≥$300M five-year forward purchase [10].
    • Management's proxy projections: Serra Verde EBITDA $33M (2026E), $556M (2027E), $735M (2028E) [4]. The deal release targets $550–650M run-rate EBITDA by end-2027 [16].
    • Deliveries to the SPV begin "early in Q4 2026" [10].
  • Why it matters: Floors take away most of the price risk that cut MP's revenue 51% and Lynas' 37% in the 2023–24 downturn [46]. If Serra Verde ramps, it alone is worth more per share than the rest of the group.
  • 😈 Devil's advocate:
    • The SPV's $500M senior debt facility is only a commitment letter. It "may not be documented, closed, or funded" [10]. The offtake can be lost if the SPV's financing is not finalised, and the SPV then holds an option to buy all of Merger Sub's equity [8].
    • The $550–650M target assumes "100% separated oxide sales" at an illustrative ~$190/kg basket from Argus in Dec 2025 [16]. On the Q2 call, when Canaccord cited ~$600M, the CFO did not confirm it; he pointed to a 6,400 t/yr run-rate by end-2027 instead [20].
    • Floors protect revenue only once tonnes are produced and separated (see Bears #2 and #3).
  • Verdict: Strong as a contract, Medium as shareholder value. The price structure is excellent. Volume, separation and the SPV's own debt are all still open.

🐂 Bull #2Western Dy and Tb trade at 4–6x Serra Verde's floors

  • The point: Ex-China heavy-rare-earth prices have split from China's and keep rising. That makes the 70% upside share in Bull #1 worth a lot.
  • Evidence:
    • S&P Global Platts (2026-09-18): Dy oxide traded at $3,250/kg and Tb oxide at $7,500/kg CIP North America. The August Platts Dy assessment was $2,300/kg [30]. Only the summary was available.
    • Fastmarkets (~2026-10-01, about events on 09-30): Tb oxide CIP global $5,000–8,000/kg, Dy oxide $2,500–3,500/kg, driven by US buyers ahead of the DFARS deadline [31]. Only the summary was available.
    • Against the floors: Dy $3,250 vs $575 (5.7x) and Tb $7,500 vs $2,050 (3.7x).
    • Dy and Tb are 32% of Serra Verde's basket value (Dy 19%, Tb 13%) at the deal's illustrative prices [16].
  • Why it matters: At these prices, Serra Verde's heavy-rare-earth revenue could run well above management's own illustrative basket. See the tailwind section for sizing.
  • 😈 Devil's advocate:
    • These are thin, trader-to-trader markets. Platts' own monthly assessment ($2,300) sat 30% below the headline trade.
    • April 2025 Chinese licensing of Dy/Tb is still in force. A US–China deal that loosened it would compress the premium. The truce runs to 2027-01-10, with no rare-earth settlement [37].
    • "Net of separation costs" means whoever separates the heavies (see Bear #3) takes a cut first.
  • Verdict: Medium The premium is real today, but it is a policy-made price and could narrow fast.

🐂 Bull #3$1.39B pro-forma cash, net cash, no going-concern doubt

  • The point: USAR is not at risk of running out of money in the next year, whatever the market does.
  • Evidence:
    • Cash was $1,530.1M at 2026-06-30, plus ~$64M restricted [1].
    • Pro forma for the Serra Verde close, cash was $1,392M: −$300M cash consideration, +$162.4M of Serra Verde cash [5].
    • Debt-like items total $533M: the DFC loan (carrying value ~$304M; SOFR+4.0%, up to 49 quarterly instalments) and the Orion royalty ($227M) [5]. That gives pro-forma net cash of ~$860M (estimate).
    • The 10-Q says cash is sufficient "for at least the next twelve months" and has no going-concern language [1].
    • Actual burn: Q2 operating cash flow −$56.9M plus ~$66–70M of capex, so free cash flow ≈ −$126M [1,19].
    • Undrawn sources: the $1.3B FFB loan (zero drawn) and ~$140M of DFC commitments [1,5].
  • Why it matters: Time to fix Serra Verde and qualify Stillwater without a distressed raise.
  • 😈 Devil's advocate:
    • Management's own plan spends much faster than the Q2 pace. The proxy shows USAR consolidated capex of $843M (2026E), $2,074M (2027E) and $959M (2028E). Its unlevered FCF is −$804M (Q2–Q4 2026), −$2,168M (2027E) and −$871M (2028E) [4].
    • Even with Serra Verde's +$494M FCF in 2027E [4], the plan exhausts $1.39B around mid-2027 unless the Commerce money and new equity arrive (estimate).
    • The FFB loan only pays against milestones (Bear #5) and carries a ~$26M/yr ticking fee while undrawn [1].
  • Verdict: Strong for 2026, Weak for the full plan. Liquidity is fine; the plan is not funded.

🐂 Bull #4Washington owns equity and has $3B+ at stake

  • The point: Several US agencies now own USAR shares or have lent into its assets. They have a strong reason not to let it fail.
  • Evidence:
    • Commerce: up to $277M of direct funding plus a $1.3B FFB loan at Treasury +150bp over 15 years. Commerce holds 16,132,790 shares and a 17,600,584-share warrant at $17.17, and keeps both even if funding is clawed back [2,1].
    • DFC: a $565M Serra Verde loan and 15.2M USAR shares registered for resale [8].
    • Department of War / US government: $750M in the offtake SPV and ≥$300M of forward purchases [10].
    • Other grants: DOE up to $19.3M for pilot separation (subject to negotiation) and the Texas Semiconductor Innovation Fund up to $14.2M [11].
  • Why it matters: It brings cheap long-dated debt, a floor-priced buyer, and political cover for the 2027 DFARS cut-off.
  • 😈 Devil's advocate:
    • Support is conditional. Commerce had disbursed $0 at 2026-06-30, and its covenants bar dividends and buybacks for 5 years and require Commerce consent for any change of control [1,2].
    • Unlike MP's deal, nothing floors the price of USAR's own magnets. The Section 232 proclamation set no tariffs or minimum import prices [44].
    • Political tail: a Senate Democrat probe of Cantor's role (July 2026) is still open [43].
  • Verdict: Strong The support is broad and has a cost, but it lowers the risk of total failure more than anything else in the file.

🐂 Bull #5DFARS 2027 ban meets a US market that is ~90% short

  • The point: From 2027-01-01, US defence systems may not use magnets with any China-linked stage. US capacity is a fraction of demand.
  • Evidence:
    • The DFARS restriction reaches every tier: mining, separation, metal and magnet [35].
    • EO 14415 (2026-07-20) replaces default waivers with Secretary-approved mitigation plans [35].
    • Reuters (2026-07-27, via Oregon Group): US NdFeB demand was ~48,000 t in 2025 against ~300 t of domestic supply. Projected US capacity of ~5,000 t by end-2026 would be about 10% of demand [35].
    • USAR reports MOUs/LOIs for 2,500 t/yr of "annual demand", 100+ prospects and 20+ customers in qualification [19].
  • Why it matters: Defence primes need qualified non-China suppliers fast. USAR plans 1,200 t/yr at Stillwater by Q1 2027 and 6,400 t/yr at Blacksburg from 2028 [11].
  • 😈 Devil's advocate:
    • Defence demand is a small slice of the 48,000 t.
    • Producers say the deadline is unachievable, which invites delay or waivers. The implementing rules are still pending (single-source date 10-18) [35].
    • MP, eVAC and Vulcan are already qualified or contracted (Bear #10).
  • Verdict: Medium The structural shortage is real, but USAR has not yet sold a magnet into it.

🐂 Bull #6Consensus already sits a third below management's plan

  • The point: The street does not model management's targets, so beating consensus does not need the full plan.
  • Evidence:
    • Consensus FY2027 revenue is $713.3M (6 analysts, range $484M–$1.01B) and EPS +$0.03 (4 analysts, range −$1.32 to +$1.07) [23].
    • Management's proxy projections for 2027E: USAR standalone revenue $308M plus Serra Verde $752M [4]. Combined ≈ $1.06B (estimate: sum of two separate projections).
    • Consensus is therefore ~33% below the combined plan. Northland models FY2027 EPS −$0.66 and FY2028 +$0.01 [48].
  • Why it matters: Part of the execution risk is already in the numbers. A Serra Verde ramp on plan would be an upside surprise.
  • 😈 Devil's advocate:
    • Consensus is built on a stale base. Price targets were last updated around 2026-09-02 [24], before the Bloomberg investigation (2026-09-23) [21] and before the stock fell another ~30%.
    • The dispersion (EPS −$1.32 to +$1.07) shows analysts do not know Serra Verde's real run-rate either.
  • Verdict: Weak Medium. The haircut helps, but consensus could still be too high (Bear #2).

🐂 Bull #7Stock trades 45% below the PIPE and the Commerce warrant

  • The point: Sophisticated money paid much more, very recently.
  • Evidence:
    • Close on 2026-10-12: $11.82, down 73% from the $43.98 high [24].
    • Other reference prices: the $1.5B PIPE at $21.50 (Jan 2026), Serra Verde holders' implied price $19.95, and the Commerce warrant strike $17.17 [2,16].
    • Short interest was 45.2M shares (~12% of ~375M outstanding; the "30% of float" figure uses a stale 148.9M float) [24,26].
    • Market cap ≈ $4.4B on ~375M shares (estimate), with ~$860M of pro-forma net cash.
  • Why it matters: Much of the bad news (overhang, funding, slips) is visible and partly priced. Any delivery could squeeze shorts.
  • 😈 Devil's advocate: Those reference prices came before the Bloomberg investigation and the funding-covenant reading. Price is not value: at $11.82 the EV is still ~5x FY2027 consensus revenue for a company whose mine produced 40 t last year (estimate). There have been no insider open-market buys (Bull #9 / Bear #6).
  • Verdict: Weak It is a sentiment point, not evidence of value.

🐂 Bull #8Non-China midstream pieces: LCM metal, Carester, Wheat Ridge

  • The point: USAR owns or part-owns some of the scarce non-China steps between oxide and magnet.
  • Evidence:
    • LCM (UK) makes rare-earth metals and alloys. Its first commercial yttrium metal pour was in April 2026 [11]. USAR bought the Hooton Park site for ~£7.0M (2026-07-02), removing the November 2026 lease expiry [1].
    • USAR holds ~13.6% of Carester. Its Caremag plant at Lacq (France) has ~5,000 t/yr of separation and recycling capacity, commissioning late 2026, and LCM Europe may buy part of its oxide [11,16].
    • Wheat Ridge (Colorado) demo plant, commissioned June 2026, includes a "third-party MREC separation" circuit [19].
  • Why it matters: These are the building blocks of a non-China route for Serra Verde output and Stillwater feed.
  • 😈 Devil's advocate:
    • LCM's gross margin was −27.2% in Q2, and four customers made up 96% of its revenue [1].
    • Caremag's heavy-oxide capacity (secondary: ~500 t/yr Dy and ~100 t/yr Tb at full ramp) is already committed in part to Brazilian Rare Earths and Lindian [45].
    • Wheat Ridge is demo scale.
  • Verdict: Weak Medium. These are real options, but none is a solution at Serra Verde's scale yet.

🐂 Bull #9A CEO who has actually run this mine

  • The point: The new CEO, Thras Moraitis (from 2026-10-01), ran Serra Verde from Jan 2023 and sat on Xstrata's executive committee. Sir Mick Davis joined the board.
  • Evidence:
    • The group was reorganised into Upstream, Midstream and Downstream units, with stated priorities of "integration, execution, capital discipline" [17].
    • Moraitis received ~0.9M of RSUs (2026-09-03 and 10-01) and holds 2.08M merger shares [18,8].
  • Why it matters: The biggest value driver is now a Brazilian mine, and the person in charge of it knows it.
  • 😈 Devil's advocate:
    • He was in charge while Serra Verde produced 40 t in 2025, wrote down $33.8M of inventory, and its auditor found two material-weakness-equivalent findings [4,7].
    • He is also a selling shareholder in the resale registration [8]. No officer or director has bought a share on the open market in 2026 [18].
  • Verdict: Weak Medium. It is the right profile, but his track record at this asset is mixed.

Bear case

🐻 Bear #1~$1.55B must be raised by end-2027, starting ~$675M by March

  • The point: The Commerce agreement requires USAR to raise large amounts of capital on fixed dates, whatever the share price.
  • Evidence:
    • From the Q2 10-Q (primary) [1]:
      • By 2027-03-31: "an additional $375.0M plus the total cash acquisition costs of Serra Verde". The cash price was $300M, so about $675M.
      • By 2027-12-31: a further $875.0M.
      • Up to $300M of each requirement may be convertible notes.
      • Requirements are reduced only by Serra Verde dividends, which USAR is told not to expect (Bear #4).
      • A revolver of up to $250M must be in place by 2027-06-30 and cash-collateralised with equity.
    • Shelf: a $1.25B universal S-3 shelf has been effective since 2026-09-15 [9].
    • Arithmetic (estimate): $1.55B at $11.82 ≈ 131M new shares, +35% on 375M. The first tranche alone ($675M) is ≈ 57M shares (+15%).
  • Why it matters: This dilution is contractual, not optional, and it falls due within 6 to 15 months at a price 45% below the January PIPE.
  • 🛡️ Counter-argument:
    • Up to $600M can be convertibles, which dilute only above a conversion premium.
    • Commerce has every reason to amend the dates rather than force a default.
    • If Serra Verde ramps, the raise could be priced off better numbers. The December lock-up expiry comes first, though.
  • Severity: High This is the most certain negative in the file.

🐻 Bear #2Serra Verde is still ramping: 40 t produced in 2025

  • The point: The "producing mine" is barely producing. Its 2026–27 targets require a ramp of roughly 100x.
  • Evidence:
    • Serra Verde's own MD&A in the proxy [4]:
      • 2025 TREO produced was 40 t (150 t in 2024). Sales were 138 t at a realised $18.33/kg, and ore processed fell 77%.
      • Revenue was $2.49M, with a $33.8M inventory impairment.
      • It "remains in the pre-operational stage" and expected "commercial operations in 2027".
      • Its prior offtake was suspended in Feb 2026, and H1 2026 revenue was $0.59M.
    • Targets: ~4,000 t/yr run-rate by end-2026, and 6,400 t/yr by end-2027 [6].
    • Bloomberg (2026-09-23, via InvestNews): investors who visited said clay/rock separation and screening problems were unsolved. USAR called these "unverified operational details" [21].
    • The USAR 10-Q, by contrast, says Serra Verde "is currently in commercial production" [1]. The two filings conflict.
  • Why it matters: Serra Verde makes up most of the $713M FY2027 consensus revenue (interpretation) and management's $556M 2027E EBITDA [4].
  • 🛡️ Counter-argument: Stage 1 debottlenecking is in advanced commissioning, and the 2025 curtailment was deliberate while the Chinese offtake was renegotiated [10,40]. A 4,000 t/yr run-rate is roughly 330 t a month, which a commissioned plant can show within one quarter.
  • Severity: High Q3 and Q4 2026 production numbers will settle this.

🐻 Bear #3No named separation route for Serra Verde output after 2026

  • The point: Serra Verde sells a mixed carbonate that someone must separate into oxides. Its Chinese processing contracts end at end-2026, and no non-China replacement has been named.
  • Evidence:
    • Reuters (2025-12-04): the Chinese deals were cut from 10 years to end-2026. Moraitis said "in a couple of years we'll have some options to separate the heavies outside of China" [40].
    • The deal release mentions "a dedicated third-party MREC separation line" and the Carester partnership, but names no operator, capacity or date [16].
    • The $550–650M EBITDA target is calculated "on the basis of 100% separated oxide sales" [16].
    • The risk factors cite "the limited availability of separation capacity outside of China" [7].
    • USAR is now on China's export-control list (2026-06-22), which bans Chinese parties from supplying it [36]. It is unclear whether this bars Chinese toll separation of its own material: not found.
    • Lynas separates heavies in Malaysia but has announced no Serra Verde deal [32].
  • Why it matters: Without separation, carbonate sells at a deep discount or not at all. The floors are oxide floors.
  • 🛡️ Counter-argument:
    • The SPV and US government have every incentive to arrange a toll route; the Lynas Texas/Malaysia capacity was built with US money.
    • Caremag starts late 2026, and Round Top's planned 8,000 t/yr separation plant is targeted for late 2028 [16].
  • Severity: High It is the single least-disclosed link in the chain.

🐻 Bear #4Serra Verde cash is locked behind DFC debt "for an extended period"

  • The point: Even if Serra Verde earns money, USAR shareholders may not see it for years.
  • Evidence:
    • Merger Sub cannot pay dividends unless all of the following hold: the optimisation project is complete, one full DFC principal instalment has been paid from offtake cash, the DSCR is ≥1.3x (historic and projected), reserves and minimum cash are funded, and there is no default [7].
    • Every distribution triggers a 50% DFC prepayment (25% on the incremental tranche), up to $465M [7].
    • The company itself says not to "expect to receive any cash dividends… for an extended period… and may never" [7].
  • Why it matters: Parent-level funding (Bear #1) cannot lean on the mine. Serra Verde EBITDA consolidates into earnings, but not into cash USAR can spend at Stillwater or Blacksburg.
  • 🛡️ Counter-argument: This is standard project-finance ring-fencing. Consolidated EBITDA still drives valuation, and refinancing the DFC loan once Serra Verde is proven would unlock it.
  • Severity: Medium High.

🐻 Bear #5Not one dollar of the Commerce package has been disbursed

  • The point: The $1.6B headline is milestone-paid, and the milestones include things USAR does not yet have.
  • Evidence:
    • At 2026-06-30: zero disbursements, zero advances, no approved request [1].
    • Milestones include feasibility studies, facility completion, capacity qualification, "securing customer purchase agreements" and target production volumes [1,2]. The exact dates are in unread exhibits.
    • The FFB loan carries a 2.0% upfront fee ($26.0M) and a 2.0%/yr ticking fee on the undrawn amount (~$26M/yr) [1].
    • Clawback triggers include missed completion deadlines [1].
    • Bloomberg (2026-09-23) reported that no milestones had been met by end-June [21]. On 2026-10-12, 24/7 Wall St tied the 6% drop to "milestone-based government funding still ahead" [27].
  • Why it matters: Until money flows, USAR pays to keep the loan open and funds its capex from cash and equity.
  • 🛡️ Counter-argument: The agreements are only four months old. The CFO expects the "first Commerce reimbursement distribution in the coming months" [19]. Fees are small against $1.39B of cash.
  • Severity: Medium A first disbursement would largely defuse it.

🐻 Bear #6Every operating promise has slipped, under a third CEO

  • The point: Guidance has a poor hit rate, and leadership keeps changing.
  • Evidence:
    • Stillwater [11,12,13,19]:
      • 2025 deck: first 1,200 t/yr "by early 2026".
      • Mar and May 2026: customer orders filled in Q2 2026.
      • Aug 2026: first sales "by year-end".
      • H1 magnet revenue was zero.
    • Commerce documents: guided for April, then May; signed 06-03 [12,13].
    • Round Top PFS (due end-Q3 2026) was folded into a DFS for Q1 2027 [11,13].
    • LCM's 3,000 t/yr end-2026 target was softened to "evaluate" [11].
    • The Serra Verde close slipped from "end of August" to 09-03 [6].
    • CEOs: Ballard → Humpton (2025-10-01) → Moraitis (2026-10-01) [17]. Humpton sold 88,391 shares on 10-02 after her RSU vest, which was a sell-to-cover [18].
    • No officer or director has bought on the open market in 2026 [18].
  • Why it matters: The funding covenants and Commerce milestones (Bears #1 and #5) are dated. More slips cost money and credibility exactly when equity must be sold.
  • 🛡️ Counter-argument: The big transactions (PIPE, Commerce, Serra Verde, TMRC, Carester) all closed roughly on time. No USAR material weakness or restatement has been reported [1].
  • Severity: High

🐻 Bear #7126M resale shares, 53M already free, 37M more on 12-02

  • The point: A third of the company is registered for resale by holders whose entry price was well above today's.
  • Evidence:
    • 126,476,950 shares are registered (33.7% of outstanding) [8]:
      • 53,150,925 have no lock-up.
      • 36,663,006 unlock on 2026-12-02.
      • 36,663,019 unlock on 2027-03-02.
    • Main holders: Serra Verde Rare Earths Ltd (Denham) 46.8M, VB (Vision Blue) 29.5M, EMG 29.2M, DFC 15.2M [8].
    • Commerce's 16.1M shares unlock on 2027-06-03 [1].
    • No filing yet shows actual selling [8,18].
  • Why it matters: The supply arrives just before and alongside the ~$675M raise.
  • 🛡️ Counter-argument: Private-equity holders like Denham and EMG usually sell in blocks rather than dumping. Vision Blue is Mick Davis's vehicle and he sits on the board. Some holders may prefer to hold through the ramp.
  • Severity: Medium High.

🐻 Bear #8Ex-China NdPr fell below $110/kg; light-rare-earth premium gone

  • The point: For NdPr, the bulk magnet input, the Western premium has disappeared.
  • Evidence:
    • Benchmark (2026-08-21): ex-China NdPr at $95/kg CIF Europe/North America, below the $110 floors, citing ex-China oversupply. China PrNd was ~$107.5/kg [29].
    • S&P Global (2026-08-18): North American sales heard below $110/kg [29,30].
    • MP guided Q3 NdPr realisation to the "high $90s" (summary, UNCONFIRMED) [33].
  • Why it matters: NdPr is 22% of Serra Verde's basket value [16]. It is floored for Phase 1, which makes the floor real money but leaves little upside. For Stillwater, a non-China magnet competes against cheap Chinese-priced NdPr content, so its premium must come from the "no China" label alone.
  • 🛡️ Counter-argument: The floor protects Serra Verde's NdPr. Heavies, the bigger part of its value at current prices, are moving the other way (Bull #2).
  • Severity: Medium

🐻 Bear #9Brazil's new law lets Brasília vet control and contracts

  • The point: Brazil passed a critical-minerals law because of this deal, and it gives a presidential council power over control changes and international contracts.
  • Evidence:
    • PL 2780/2024 passed the Chamber on 05-06 and the Senate on 09-02, and Lula signed it without vetoes on 2026-09-16 [38].
    • It creates the Cimce council under the Presidency. Cimce can "homologate" transfers of corporate control in critical-mineral holders and international contracts, and it links incentives (R$2B Fgam fund, R$5B of tax credits) to domestic processing [38].
    • The bill's author said the Serra Verde sale accelerated it. Lula attacked "traidores da pátria" at the signing [38].
    • The Rede party's STF petition (ADPF, filed 2026-04-24) against the sale has no ruling found [39].
    • Whether the law applies to Serra Verde retroactively: not found.
  • Why it matters: An export-conditioning or domestic-processing rule would hit the carbonate-export model and Phase 2.
  • 🛡️ Counter-argument: The deal closed after Senate passage, and Brazil wants investment and Fgam-type projects. A country-wide election (Oct 2026) may soften or harden this. ANM said no prior approval was needed [39].
  • Severity: Medium

🐻 Bear #10Rivals already hold binding magnet contracts and floors

  • The point: In US magnets, USAR is late and has no contract.
  • Evidence:
    • MP Materials [33]:
      • Q2 2026 revenue was $108.5M, including $16.5M of magnetics.
      • DoD $110/kg NdPr floor plus 100% offtake of its 10X magnets.
      • Apple agreement.
      • $1.45B of cash.
    • eVAC Sumter: 2,000 t/yr with a GM contract; being bought by Energy Fuels (~$1.9B, early 2027) [34].
    • Vulcan Elements supplies Army drones [35].
    • Lynas: FY26 NPAT A$222.4M, a 12-year JARE NdPr deal with a $110 floor, and Malaysian heavy separation [32].
    • USAR: "no definitive off-take or sales agreements" for magnets [7].
  • Why it matters: Defence primes qualify a supplier once. Being third or fourth means smaller allocations.
  • 🛡️ Counter-argument: None of these rivals owns a heavy-rare-earth mine with Dy/Tb floors. The 48,000 t US gap is big enough for several winners [35].
  • Severity: Medium High.

🐻 Bear #11Round Top's grade is ~0.03% TREO; critics call it uneconomic

  • The point: USAR's US mine, which takes $132M + $550M of the Commerce package [1], may not be economic.
  • Evidence:
    • CSIS (Feb 2026, via secondary): ~0.03% TREO grade, Dy ~30 ppm vs ~400 ppm at Mt Weld. That means 100–150x more rock than Mt Weld per unit of output [21].
    • Bloomberg (2026-09-23): Jon Hykawy called it "laughably bad". Former officials said Pentagon reviews since the 1990s found no economic grade [21,22].
    • The only published economics are a 2019 PEA. The DFS is due Q4 2026, with publication in Q1 2027 [11].
  • Why it matters: A weak DFS could delay or claw back Commerce money tied to Round Top and remove the domestic heavy-rare-earth story.
  • 🛡️ Counter-argument: ~72% of the oxide is heavy rare earths, a rare mix (Wedbush). Recent drilling assays "matched expected resource grades" [19]. Serra Verde makes Round Top non-essential.
  • Severity: Medium It is mainly a Commerce-milestone and credibility risk.

🐻 Bear #12China blacklist, MP lawsuit and a Senate probe hang over it

  • The point: Legal and geopolitical exposures each add tail risk.
  • Evidence:
    • China's export-control list since 2026-06-22. The company says it "has had and is expected to continue to have an adverse impact" on sourcing [7,36].
    • MP v. USAR (Texas Business Court, filed 2026-05-22) alleges trade-secret misappropriation, including grain-boundary-diffusion know-how [41,1]:
      • MP seeks an injunction and damages above $5M.
      • USAR denied "each and every" allegation on 2026-06-22 [42].
      • No hearing or injunction motion was found, and no accrual has been booked.
    • The Senate probe of Cantor/Lutnick ties is open, with letters sent in July 2026 [43].
  • Why it matters: An injunction on GBD would hit Stillwater's high-temperature grades, the ones defence buyers need.
  • 🛡️ Counter-argument: GBD is widely practised in Japan and China. The damages sought are small. The China listing mostly formalises what was already happening.
  • Severity: Low Medium.

Tailwinds NOT in the forecasts

FY2027 consensus: revenue $713.3M, EPS +$0.03 [23]. Because EPS is near zero, "% vs consensus EPS" is meaningless. EPS impacts are given in $/share on ~375M shares, pre-tax (USAR has NOLs; estimate). Consensus share counts are not disclosed.

  • Heavy-rare-earth upside above the floors reaches the P&L in 2027 — 35% chance in FY2027

    • Impact on revenue: +$140M, +20% vs consensus FY2027.
    • Impact on EPS: +$0.37 per share in FY2027 (pre-tax; not meaningful as % of +$0.03 consensus).
    • How we got the number (estimate):
      • Illustrative basket $190/kg × 4,400 t ≈ $836M. Dy 19% ≈ 144 t and Tb 13% ≈ 27 t at ~$1,100 and ~$4,000/kg (presentation spot; garbled labels, verify).
      • At today's $3,250/$7,500, extra gross value ≈ 144×$2,150 + 27×$3,500 ≈ $405M. 70% share ≈ $283M.
      • Assume half survives separation costs and premium fade: ≈ $140M ÷ 375M ≈ +$0.37.
    • Expected value: 0.35 × $0.37 ≈ +$0.13 per share.
    • What it is: Serra Verde realises the 70% upside share on Dy and Tb at current Western prices.
    • Why it's not priced in: Analysts work off management's Dec-2025 basket. No published analyst work on realised prices or the upside formula was found. Consensus FY2027 is ~33% below management's own plan [4,23].
    • Trigger to watch: Realised SPV prices disclosed with Q3/Q4 results. Platts and Fastmarkets Dy/Tb CIP prints.
    • Why this probability: The prices exist today. But volume (Bear #2), separation (Bear #3) and the premium's survival through 2027 all have to line up.

  • A binding multi-year defence or auto magnet contract for Stillwater — 40% chance by end-2027

    • Impact on revenue: +$72M, +10% vs consensus FY2027.
    • Impact on EPS: +$0.06 per share FY2027.
    • How we got the number (estimate): 600 t × $120/kg (assumed non-China magnet price; not sourced) = $72M × 30% incremental margin ≈ $22M ÷ 375M ≈ +$0.06.
    • Expected value: 0.40 × $0.06 ≈ +$0.02.
    • What it is: One of the 2,500 t/yr of MOUs turns into a named, binding offtake after the 2027-01-01 DFARS cut-off.
    • Why it's not priced in: The company has no definitive offtakes [7]. Magnet revenue in consensus is unknown but likely small.
    • Trigger to watch: An 8-K naming a customer and volumes; AS9100 certification; a Commerce milestone tied to "customer purchase agreements" being met.
    • Why this probability: Demand pressure is real, but USAR is behind MP, eVAC and Vulcan in qualification. A contract would matter more for the share price than for EPS.

  • China re-tightens controls after the truce lapses on 2027-01-10 — 25% chance within 12 months

    • Impact on revenue: not quantifiable separately. It would widen the Western Dy/Tb premium already counted in the first tailwind and lift ex-China NdPr back above $110.
    • Impact on EPS: not quantifiable. It overlaps the first tailwind and would also squeeze LCM's input costs.
    • How we got the number: no separate basis; the effect works through prices already sized above.
    • Expected value: n/a.
    • What it is: The suspended Oct-2025 controls return.
    • Why it's not priced in: The September and October sell-offs priced a thaw [27,28].
    • Trigger to watch: MOFCOM notices before 2027-01-10.
    • Why this probability: Both sides have extended twice. Renewal is the base case, but the summit settled nothing on rare earths [37].

  • A price floor or offtake for USAR's own magnets (MP-style) — 15% chance within 36 months

    • Impact on revenue: +$145M, +20% vs consensus FY2027 if it arrived in time; more likely FY2028.
    • Impact on EPS: +$0.12 per share.
    • How we got the number (estimate): 1,200 t Phase 1 × $120/kg = $145M × 30% margin ≈ $43M ÷ 375M ≈ +$0.12.
    • Expected value: 0.15 × $0.12 ≈ +$0.02.
    • What it is: A DoW or Commerce offtake with a price floor on Stillwater output, or Section 232 minimum import prices.
    • Why it's not priced in: The Commerce deal has no floor [2]. Proclamation 11001 reserved minimum import prices but set none [44].
    • Trigger to watch: A Commerce/USTR 232 follow-up, or a DoW procurement award.
    • Why this probability: Washington has used floors for MP and Serra Verde. A third, for an unqualified magnet maker, is a stretch.

Headwinds NOT in the forecasts

  • Serra Verde 2027 output or separation falls ≥30% short of plan — 50% chance in FY2027

    • Impact on revenue: −$200M, −28% vs consensus FY2027.
    • Impact on EPS: −$0.27 per share FY2027.
    • How we got the number (estimate): Serra Verde ≈ $650M of the $713M consensus (my reading) × 30% shortfall ≈ −$200M × ~50% decremental EBITDA margin ≈ −$100M ÷ 375M ≈ −$0.27.
    • Expected value: 0.50 × −$0.27 ≈ −$0.13 per share.
    • What it is: The ramp from 40 t (2025) to a 4,000–6,400 t/yr run-rate stalls, or carbonate cannot be separated outside China after end-2026.
    • Why it's not priced in: Consensus assumes a ramp, and no analyst has published a separation-route analysis.
    • Trigger to watch: Q3/Q4 production tonnes; a named toll-separation contract; the SPV's $500M debt closing.
    • Why this probability: Serra Verde's own filing said commercial operations in 2027, separation is unnamed, and visitors reported processing problems. A coin-flip is fair given US-government help on the other side.

  • Equity and converts of ≥$675M by 2027-03-31, at or below today's price — 80% chance by 2027-03-31

    • Impact on revenue: none.
    • Impact on EPS: −15% on every FY2027 per-share figure. Consensus EPS +$0.03 → ~+$0.026, which is not meaningful in dollars. On the proxy's 2027E combined EBITDA ($556M − $35M = $521M; estimate), EBITDA per share falls from $1.39 to ~$1.21 (−$0.18).
    • How we got the number (estimate): $675M ÷ $11.82 ≈ 57M shares; 375M → 432M (+15%). $521M ÷ 432M ≈ $1.21 vs $1.39.
    • Expected value: 0.80 × −$0.18 ≈ −$0.14 EBITDA per share.
    • What it is: The first contractual Commerce raise (Bear #1). The second, $875M by end-2027, would add a similar step again.
    • Why it's not priced in: Price targets average ~$35 and were set around 09-02 [24]. Consensus share counts appear to be pre-raise (not confirmed).
    • Trigger to watch: An ATM agreement or underwritten deal off the $1.25B shelf, a convertible issue, or a Commerce amendment.
    • Why this probability: It is a contractual requirement with a fixed date. The 20% covers an amendment, a waiver, or a raise well above today's price.

  • Stillwater misses year-end first sales and the 600 t/yr run-rate — 55% chance by Q1 2027

    • Impact on revenue: −$30M, −4% vs consensus FY2027.
    • Impact on EPS: −$0.03 per share.
    • How we got the number (estimate): Assume ~$60M of magnet revenue in FY2027 consensus (not disclosed). Half slips ≈ −$30M × 35% margin ≈ −$10.5M ÷ 375M ≈ −$0.03.
    • Expected value: 0.55 × −$0.03 ≈ −$0.02. The damage is mostly to credibility and the Commerce milestones, not EPS.
    • What it is: The third slip of the first-sales date.
    • Why it's not priced in: Guidance "by year-end" was reiterated in August [19].
    • Trigger to watch: Q3 results (~2026-11-06 to 11-09, unconfirmed).
    • Why this probability: It slipped twice already, the LCM feedstock delay was admitted in writing [7], and the plant had 140 of 200 targeted staff in August [19].

  • Brazil applies the new law to restrict raw MREC exports or Phase 2 — 15% chance within 36 months

    • Impact on revenue: −$150M, −21% vs consensus FY2027 (more likely FY2028).
    • Impact on EPS: −$0.20 per share.
    • How we got the number (estimate): a ~25% export or volume restriction on ~$650M ≈ −$160M × 50% margin ≈ −$80M ÷ 375M ≈ −$0.20.
    • Expected value: 0.15 × −$0.20 ≈ −$0.03.
    • What it is: Cimce uses its homologation powers, or regulations link exports to domestic processing [38].
    • Why it's not priced in: The law passed in September with little US coverage; analysts have not mentioned it.
    • Trigger to watch: Implementing decrees; a Cimce ruling on Serra Verde; the STF ADPF.
    • Why this probability: The rhetoric is hostile, but Brazil courts US capital, and a retroactive hit on a closed deal is legally harder.

  • MP wins an injunction on Stillwater's GBD process — 10% chance within 18 months

    • Impact on revenue: −$20M, −3% vs consensus FY2027.
    • Impact on EPS: −$0.02 per share.
    • How we got the number (estimate): ~170 t of high-temperature grades × $120/kg ≈ $20M × 35% ≈ $7M ÷ 375M ≈ −$0.02.
    • Expected value: 0.10 × −$0.02 ≈ −$0.002.
    • What it is: A court order bars use of the contested know-how.
    • Why it's not priced in: The market treats the suit as noise.
    • Trigger to watch: A temporary-injunction motion or hearing on the Texas Business Court docket.
    • Why this probability: Trade-secret injunctions are rare, and none has been sought on record so far.

Investment Risk Score: 69/100 (high)

# Factor Weight Measured value Score Why
1 Liquidity & funding runway 15% Pro-forma cash $1.39B; Q2 FCF ≈ −$126M → >10 qtrs at current pace; management plan UFCF Q2–Q4'26 −$804M and 2027 −$2.17B (+$0.49B Serra Verde) → ~12–15 months on plan; $0 Commerce disbursed 64 The plan runway is 12–36 months, at the short end. No going-concern language, but contractual raises are due 03-27 and 12-27.
2 Leverage 10% Net cash ≈ $860M pro forma; debt-like ≈ $533M (DFC ~$304M at SOFR+4%, Orion royalty ~$227M); FFB $1.3B to be drawn; negative EBITDA 35 Net cash today. Planned borrowing is large while EBITDA is negative, so the score sits above the 0–20 band.
3 Dilution 10% Shares 148.1M (12/31/25) → ~375M (09-26) = +153%; required raises ~$1.55B ≈ +35% at $11.82; DOC warrant 17.6M 92 Far above the +25% anchor, with more dilution contractually required.
4 Profitability 10% Q2 revenue $5.8M, gross margin −27.2%, operating loss −$46.3M; Serra Verde H1 revenue $0.59M, operating loss −$39.8M 88 Deep losses; the main asset is pre-commercial.
5 Revenue quality & concentration 10% LCM: four customers = 96% of Q2; Serra Verde: one buyer (SPV); no binding magnet contract 75 Effectively one asset and one buyer for future revenue. The buyer is US-government backed, which keeps it below 81.
6 Execution & track record 10% First magnet sales slipped Q2 → YE; Commerce docs April → June; PFS folded; LCM 3,000 t softened; 3 CEOs in 13 months; Serra Verde 2 MW-equivalent findings 78 Serial slips. No USAR material weakness, so not 81+.
7 Competitive position & disruption 10% MP (DoD floor + offtake, $1.45B cash), eVAC (GM), Vulcan, Lynas ahead in magnets/separation; Serra Verde = rare non-China ionic-clay heavy source 58 Contested in magnets, strong in heavy feed.
8 Cyclicality & commodity/macro sensitivity 10% Rare-earth prices; 2023–24 MP revenue −51%, Lynas −37%; ex-China NdPr $95/kg (Aug); Phase-1 floors protect Serra Verde only 62 Highly commodity- and policy-linked. Floors cover the contracted mine output, not LCM or magnets.
9 Regulatory, legal & geopolitical 5% China export-control list; Brazil Cimce law (09-16) + STF petition; MP suit; Senate probe; Commerce clawback/debarment 72 Large pending exposures with China and Brazil. Not existential on current evidence.
10 Valuation & expectations 10% Market cap ≈ $4.4B, EV ≈ $3.6B ≈ 5x FY27 / 49x FY26 consensus revenue; short 45.2M shares (~12% of shares); average target ~$35 is stale 70 Priced for a ramp that has not shown up yet. Heavy shorting and a fragile story.

Overall: weighted average = 0.15×64 + 0.10×35 + 0.10×92 + 0.10×88 + 0.10×75 + 0.10×78 + 0.10×58 + 0.10×62 + 0.05×72 + 0.10×70 = 9.6 + 3.5 + 9.2 + 8.8 + 7.5 + 7.8 + 5.8 + 6.2 + 3.6 + 7.0 = 69.0 → 69.

Change vs last: +4 (65 under the old 6-area method). What moved:

  • The 10-factor method weighs dilution and profitability separately, and both are extreme.
  • New facts raised execution and valuation: the exact ~$675M + $875M raise covenants, Serra Verde's 40 t of 2025 output and "pre-operational" status, and $0 of Commerce disbursements.
  • Net cash keeps leverage low.

Ranking check: At 69, USAR ties IREN (69) and sits just above IPX (67). Like IPX it has no profitable product and depends on milestone-paid government money. But its contractual raises, 33.7% resale overhang and unproven mine make it riskier. Its $1.39B of pro-forma cash and floored offtake keep it below WYFI (70), whose own 10-Q warns that resources "may not be sufficient".

Conclusion

  • What you are really betting on: That Serra Verde becomes a real 4,000–6,400 t/yr heavy-rare-earth producer with a non-China separation route before USAR has to sell ~$1.55B of new equity. In that case, US policy (floors, DFARS, Commerce) turns a mine plus a metal-maker into a cash-generating, strategic asset. The magnet plant is the name on the door, but the mine and the financing decide the per-share outcome.
  • The 2–3 things that will decide the outcome:
    1. Serra Verde's production tonnes and a named separation route for 2027 carbonate.
    2. The size, price and form (equity vs convertible) of the raise due by 2027-03-31.
    3. Whether Commerce money starts to flow, which depends on milestones including customer contracts.
  • Signposts:
    • Confirm bull:
      • Serra Verde reports ≥300 t/month TREO, or a 4,000 t/yr run-rate, by Q4 2026.
      • A named toll-separation agreement (Lynas, Caremag, Solvay or another non-China route).
      • The first Commerce/FFB disbursement and the SPV's $500M debt closing.
      • Realised Dy/Tb prices disclosed well above the floors.
      • The raise is done as convertibles or above ~$15.
    • Break bull:
      • Q3/Q4 Serra Verde output stays in the tens of tonnes.
      • No magnet revenue in Q4 2026.
      • An equity raise below ~$10, or a request to amend the Commerce raise dates.
      • Cimce or a Brazilian decree conditions MREC exports.
      • The Round Top DFS confirms sub-economic grade.
  • New since the last report:
    • The funding covenant is now exact: ~$675M by 2027-03-31 (incl. $300M Serra Verde cash cost), $875M by 2027-12-31, up to $300M of each as converts, and a cash-collateralised $250M revolver by 2027-06-30. A $1.25B shelf has been effective since 09-15.
    • Serra Verde is far earlier-stage than the old report assumed: 40 t produced and $2.49M of revenue in 2025, "pre-operational", commercial operations expected in 2027. Its dividends to USAR are locked behind DFC tests and a 50% prepayment sweep.
    • Brazil signed a critical-minerals law on 09-16 that gives a presidential council power over control transfers and contracts. Ex-China NdPr fell below $110/kg in August, while Dy/Tb rose to $3,250/$7,500/kg.

Open questions & data gaps

  • Separation route for Serra Verde carbonate from 2027: nothing named in any filing. Whether Chinese toll separation continues after end-2026, and whether China's 06-22 listing allows it, is unknown. This matters most of all for the 2027 revenue verdict.
  • Commerce milestone schedule and dates: in exhibits to the DFA/LGA that were not read. Unknown whether the $50M PIPE excess carries forward, and how big the revolver collateral is. This matters for the size of the raise and the clawback risk.
  • Serra Verde Q3 2026 production: no interim data. The 10-Q says "currently in commercial production" while Serra Verde's own MD&A says "pre-operational". The Q3 10-Q (~early Nov) is the first consolidated look.
  • Verbatim Q1/Q2 call Q&A: only summaries (MarketBeat, Investing.com) were available. Management quotes in this report are therefore from releases or summaries.
  • Consensus split between Serra Verde, magnets and LCM, and FY2028 consensus: not found (paywalled models). This matters for tailwind/headwind sizing; my split is an estimate.
  • Bloomberg 2026-09-23 investigation: read only via Portuguese and Romanian republications. The CSIS Round Top paper was not opened.
  • The illustrative basket's Dy/Tb prices from the April presentation were garbled in extraction (~$1,100 and ~$4,000/kg). The first tailwind's sizing depends on them.
  • Blacksburg lease: the landlord could terminate if financing did not close within ~90 days of 06-01. No update found, although groundbreaking was held on 09-09/09-16.
  • Selling by Serra Verde holders (Denham, EMG, VB, DFC): no filings yet. 13G/13D amendments or the Q3 10-Q holder data would show it.
Sources
  1. Form 10-Q Q2 2026 — USA Rare Earth / SEC EDGAR — 2026-08-10 — https://www.sec.gov/Archives/edgar/data/1970622/000197062226000057/usar-20260630.htm
  2. Form 8-K, Commerce definitive agreements (Items 1.01, 2.03, 3.02) — SEC EDGAR — 2026-06-03 — https://www.sec.gov/Archives/edgar/data/1970622/000121390026064453/ea0293402-8k_usarare.htm
  3. Commerce definitive agreements press release (8-K Ex. 99.1) — SEC EDGAR — 2026-06-03 — https://www.sec.gov/Archives/edgar/data/1970622/000121390026064453/ea029340201ex99-1.htm
  4. DEFM14A merger proxy (Serra Verde MD&A, projections, government financing) — SEC EDGAR — 2026-07-24 — https://www.sec.gov/Archives/edgar/data/1970622/000121390026081091/ea0290028-06.htm
  5. Unaudited pro forma financial information (8-K Ex. 99.3) — SEC EDGAR — 2026-09-04 — https://www.sec.gov/Archives/edgar/data/1970622/000121390026097399/ea030400101ex99-3.htm
  6. Serra Verde combination completed (8-K Ex. 99.1) — SEC EDGAR — 2026-09-04 — https://www.sec.gov/Archives/edgar/data/1970622/000121390026097399/ea030400101ex99-1.htm
  7. Supplemental risk factors (8-K Ex. 99.1) — SEC EDGAR — 2026-09-15 — https://www.sec.gov/Archives/edgar/data/1970622/000121390026100190/ea030552601ex99-1.htm
  8. 424B3 resale prospectus (126,476,950 shares; lock-ups; selling holders; offtake terms) — SEC EDGAR — 2026-09-17 — https://www.sec.gov/Archives/edgar/data/1970622/000121390026100977/ea0305798-424b3_usarare.htm
  9. Form S-3 $1.25B universal shelf — SEC EDGAR — 2026-09-04 (effective 2026-09-15) — https://www.sec.gov/Archives/edgar/data/1970622/000121390026097754/ea0304147-s3_usarare.htm
  10. SPV capitalisation 8-K and Serra Verde release (Ex. 99.1/99.2) — SEC EDGAR — 2026-08-24 — https://www.sec.gov/Archives/edgar/data/0001970622/000121390026092810/ea030292401ex99-2.htm
  11. Q2 2026 earnings release (8-K Ex. 99.1) — SEC EDGAR — 2026-08-10 — https://www.sec.gov/Archives/edgar/data/0001970622/000197062226000056/exhibit991-earningsrelease.htm
  12. Q1 2026 earnings release (8-K Ex. 99.1) — SEC EDGAR — 2026-05-13 — https://www.sec.gov/Archives/edgar/data/0001970622/000197062226000034/exhibit991-earningsrelease.htm
  13. Q4/FY2025 earnings release (8-K Ex. 99.1) — SEC EDGAR — 2026-03-30 — https://www.sec.gov/Archives/edgar/data/1970622/000197062226000019/exhibit991-earningsrelease.htm
  14. Form 10-K FY2025 — SEC EDGAR — 2026-03-30 — https://www.sec.gov/Archives/edgar/data/1970622/000197062226000021/usar-20251231.htm
  15. Form 10-Q Q1 2026 — SEC EDGAR — 2026-05-14 — https://www.sec.gov/Archives/edgar/data/0001970622/000197062226000038/usar-20260331.htm
  16. Serra Verde acquisition release and investor presentation (8-K Ex. 99.1/99.3) — SEC EDGAR — 2026-04-20 — https://www.sec.gov/Archives/edgar/data/0001970622/000121390026045339/ea028691001ex99-3.htm
  17. CEO transition releases (8-K) — SEC EDGAR — 2026-07-20 and 2026-10-01 — https://www.sec.gov/Archives/edgar/data/1970622/000121390026105454/ea030718301ex99-1.htm
  18. Form 4 and Form 144 filings, CIK 1970622 (2026-03-19 to 2026-10-05) — SEC EDGAR — 2026-10-05 — https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=1970622&type=4
  19. USA Rare Earth Q2 earnings call highlights (summary) — MarketBeat — 2026-08-10 — https://www.marketbeat.com/instant-alerts/usa-rare-earth-q2-earnings-call-highlights-2026-08-10/
  20. Earnings call transcript: USA Rare Earth Q2 2026 — Investing.com — 2026-08-10 — https://www.investing.com/news/transcripts/earnings-call-transcript-usa-rare-earth-q2-2026-revenue-miss-sparks-selloff-93CH-4850510
  21. Desafio da empresa dos EUA que comprou a Serra Verde (Bloomberg content) — InvestNews — 2026-09-23 — https://investnews.com.br/negocios/desafio-empresa-eua-comprou-serra-verde-terras-raras/
  22. Magnet maker attracted $1.6 billion (Bloomberg; summary only) — Advisor Perspectives — 2026-09-23 — https://www.advisorperspectives.com/articles/2026/09/23/magnet-maker-attracted-1-6-billion-lutnick
  23. USAR analysis and estimates — Yahoo Finance — 2026-10-12 — https://finance.yahoo.com/quote/USAR/analysis/
  24. USAR forecast and statistics — StockAnalysis — 2026-10-12 — https://stockanalysis.com/stocks/usar/forecast/
  25. USAR analyst ratings history (Roth, Needham cuts) — MarketBeat — 2026-07-22 — https://www.marketbeat.com/stocks/NASDAQ/USAR/forecast/
  26. USAR short interest — Fintel — 2026-10-13 — https://fintel.io/ss/us/usar
  27. USA Rare Earth drops 6% with milestone-based government funding still ahead — 24/7 Wall St — 2026-10-12 — https://247wallst.com/investing/2026/10/12/usa-rare-earth-drops-6-with-milestone-based-government-funding-still-ahead-of-it-mp-materials-falls-6-critical-metals-holds-steady/
  28. Rare earth stocks tumble on U.S.-China thaw hopes — 24/7 Wall St — 2026-09-10 — https://247wallst.com/investing/2026/09/10/rare-earth-stocks-tumble-on-u-s-china-thaw-hopes-usa-rare-earth-sinks-4-mp-materials-drops-5-critical-metals-slips/
  29. Ex-China NdPr prices slip below $110/kg price floor as oversupply builds — Benchmark Mineral Intelligence — 2026-08-21 — https://source.benchmarkminerals.com/article/ex-china-ndpr-prices-slip-below-110-kg-price-floor-as-oversupply-builds
  30. Rare earths: higher-priced dysprosium, terbium oxide trade the focus in Sept (summary only) — S&P Global Commodity Insights — 2026-09-18 — https://www.spglobal.com/energy/en/news-research/latest-news/metals/091826-rare-earths-higher-priced-dysprosium-terbium-oxide-trade-the-focus-in-sept-cif-north-america-rees
  31. US demand drives sharp increases in CIP global terbium, dysprosium prices (summary only) — Fastmarkets — ~2026-10-01 — https://www.fastmarkets.com/?p=137339
  32. Lynas FY2026 results — ASX announcement — 2026-08-26 — https://announcements.asx.com.au/asxpdf/20260826/pdf/0737tbnynjls6y.pdf
  33. MP Materials reports second quarter 2026 results — MP Materials — 2026-08-06 — https://mpmaterials.com/news/mp-materials-reports-second-quarter-2026-results
  34. Energy Fuels to acquire VAC for $1.9B — Metal Tech News — 2026-06-24 — https://www.metaltechnews.com/story/2026/06/24/tech-metals/energy-fuels-to-acquire-vac-for-19b/2809.html
  35. US suppliers not ready for Pentagon's 2027 ban on China-sourced magnets (Reuters 2026-07-27) — The Oregon Group — 2026-07-30 — https://theoregongroup.com/commodities/rare-earths/us-suppliers-not-ready-for-pentagons-2027-ban-on-china-sourced-magnets-and-minerals/
  36. China imposes export control and government procurement restrictions on designated US companies — Arnold & Porter — 2026-07 — https://arnoldporter.com/en/perspectives/advisories/2026/07/china-imposes-export-control-and-government-procurement-restrictions-on-designated-us-companies
  37. Chinese rare earth exports: Trump–Xi truce extended — Rinnovabili — 2026-09-25 — https://www.rinnovabili.net/policy-and-affairs/environmental-policies/chinese-rare-earth-exports-trump-xi-truce/ ; Bessent announces two-month extension of US-China trade truce — The Star — 2026-09-23 — https://www.thestar.com.my/aseanplus/aseanplus-news/2026/09/23/bessent-announces-two-month-extension-of-us-china-trade-truce
  38. Lula sanciona lei que institui Política Nacional de Minerais Críticos — DGABC — 2026-09-16 — https://www.dgabc.com.br/Noticia/4347341/lula-sanciona-lei-que-institui-politica-nacional-de-minerais-criticos- ; Lula centraliza no Planalto decisões sobre minerais críticos — Poder360 — 2026-09 — https://www.poder360.com.br/poder-governo/lula-centraliza-no-planalto-decisoes-sobre-minerais-criticos/ ; Lula volta a atacar "traidores da pátria" — Eixos — 2026-09-16 — https://eixos.com.br/politica/lula-volta-a-atacar-traidores-da-patria-ao-sancionar-politica-de-minerais-criticos/
  39. AGU pede que STF rejeite ação contra venda da Serra Verde — Times Brasil — 2026-06 — https://timesbrasil.com.br/brasil/terras-raras-agu-pede-que-stf-rejeite-acao-contra-venda-da-mineradora-serra-verde/
  40. Serra Verde cuts short China offtake deals, approached Western firms (Reuters) — Kitco — 2025-12-04 — https://www.kitco.com/news/off-the-wire/2025-12-04/serra-verde-cuts-short-china-offtake-deals-approached-western-firms
  41. MP Materials accuses USA Rare Earth of tech theft — The Northern Miner — 2026-05 — https://www.northernminer.com/news/mp-materials-accuses-usa-rare-earth-of-tech-theft/1003891523/
  42. USA Rare Earth rebuts MP Materials lawsuit (summary via Bitget) — Bloomberg — 2026-06-22 — https://www.bloomberg.com/news/articles/2026-06-22/usa-rare-earth-rebuts-mp-materials-lawsuit-as-rivalry-deepens
  43. Van Hollen, Warren, Wyden, Lofgren lead new probe into Cantor Fitzgerald, USAR — US Senate — 2026-07 — https://www.vanhollen.senate.gov/news/press-releases/van-hollen-warren-wyden-lofgren-lead-new-probe-into-cantor-fitzgerald-usar-after-16-billion-trump-administration-deal
  44. Section 232 critical minerals results and negotiations — Covington — 2026-01 — https://www.cov.com/en/news-and-insights/insights/2026/01/trump-administration-announces-results-of-critical-minerals-investigation-under-section-232-and-directs-us-officials-to-initiate-negotiations
  45. Terres rares: Carester sécurise 216 millions d'euros (summary only) — Le Journal des Entreprises — 2025 — https://www.lejournaldesentreprises.com/article/terres-rares-le-lyonnais-carester-securise-216-millions-deuros-et-lance-la-construction-de-son-usine-2115677
  46. Lynas posts 75% profit drop amid softening China demand — Capital Brief — 2024 — https://www.capitalbrief.com/briefing/lynas-posts-75-profit-drop-amid-softening-china-demand-297eceb7-b147-4334-bae0-4ff84952e7f4/ ; MP Materials reports second quarter 2024 results — MP Materials — 2024-08 — https://mpmaterials.com/articles/mp-materials-reports-second-quarter-2024-results
  47. Jefferies initiates USA Rare Earth at Buy, $21 target (summary) — Midas — 2026-09-02 — https://www.getmidas.com/midasin-kulaklari/jefferies-usa-rare-earthe-21-dolar-hedef-fiyatla-al-tavsiyesiyle-kapsam-baslatti-p-729623
  48. Analysts issue forecasts for USA Rare Earth FY2028 earnings (Northland) — MarketBeat — 2026-08-14 — https://www.marketbeat.com/instant-alerts/analysts-issue-forecasts-for-usa-rare-earth-fy2028-earnings-2026-08-14/
  49. Blacksburg lease and incentives (8-K) — SEC EDGAR — 2026-06-02 — https://www.sec.gov/Archives/edgar/data/1970622/000121390026063832/ea0293127-8k_usarare.htm
Disclaimer. AI-generated research and podcast (Claude by Anthropic, voices by Google Gemini) for educational purposes — not financial advice and not a recommendation to buy or sell. Willem (MR.INVEST) owns this stock. Check the company’s own filings before making any decision.
← All Bulls vs Bears reports