Lynas Rare Earths

Phase A — Understand the business

Lens 1 · Company Overview

Lynas is a vertically integrated rare-earths miner-processor: it mines rare-earth-bearing ore at Mt Weld (Western Australia), concentrates it, performs cracking & leaching at Kalgoorlie (WA), and does the value step — solvent-extraction separation into individual oxides — at the Lynas Advanced Materials Plant (LAMP) in Gebeng/Kuantan, Malaysia. It is the largest producer of separated rare-earth materials outside China and, since 2025, the only commercial producer of separated heavy rare earths (dysprosium, terbium) outside China.

What it actually sells: separated rare-earth oxides. The economic engine is the NdPr oxide family (neodymium-praseodymium) — the input to NdFeB permanent magnets used in EV traction motors, wind-turbine generators, robotics and defence actuators. NdPr is ~the entire margin story; the newly separated Dy/Tb are lower-volume but strategically precious (they raise a magnet's heat resistance and are ~99–100% China-controlled).

Scale (H1 FY2026, six months to 31 Dec 2025): revenue A$413.7M, NdPr production 3,407 t (+15% YoY), total REO 6,375 t. Run-rate NdPr capacity ~7,200 tpa currently, with facility upgrades taking nameplate toward ~10,500 tpa and the funded target of 12,000 tpa at full Mt Weld ramp.

Customers / contract structure: the anchor is JARE (Japan Australia Rare Earths B.V.), a JOGMEC + Sojitz JV that has held long-term offtake since Lynas's near-death recapitalisation. In March 2026 that deal was extended to 2038 with a firm offtake of 5,000 t NdPr/yr at a US$110/kg price floor, now covering light and heavy oxides. On the US side, Lynas signed a binding Letter of Intent with the US Department of War (16 Mar 2026), ~US$96M over four years for light + heavy oxide purchases. Contracts are a mix of take-or-pay-flavoured offtake (JARE) and government-backed floors — the structure is deliberately built to escape China's spot-price whipsaw, though most of the volume still clears at market.

Lens 2 · Supply Chain

Named end-to-end map (names or it didn't happen):

Upstream (inputs into Lynas):

• Ore: self-supplied from Mt Weld (WA) — a monazite/bastnäsite carbonatite Lynas owns outright; >20-year reserve life at the expanded 12,000-tpa NdPr rate. This is the key structural advantage vs peers who must buy concentrate.
• Reagents / consumables: hydrochloric & sulfuric acid, caustic soda for cracking-leaching-separation (generic chemical supply; not disclosed by name).
• Energy: historically diesel + grid at Mt Weld/Kalgoorlie; now 95% renewable penetration at Mt Weld via a new hybrid renewable power station (870,000+ litres diesel saved).

Midstream (Lynas itself — the chokepoint it owns):

• Mt Weld concentrate → Kalgoorlie cracking & leaching plant (WA, commissioned to reduce reliance on doing that step in Malaysia) → LAMP Malaysia separation → individual NdPr / La / Ce / SEG-Dy-Tb oxides.
• The new heavy-REE separation circuit at LAMP (commissioned Mar-quarter 2025) can separate up to 1,500 tpa of mixed SEGH (samarium, europium, gadolinium, holmium, dysprosium, terbium).
• Planned US step: Seadrift, Texas processing (DoD-funded) — see chokepoint note below.

Downstream (buyers):

• JARE / JOGMEC / Sojitz (Japan) — largest long-term offtaker, ~30% of Japan's heavy-REE demand covered.
• US Department of War (via the US$96M LOI).
• Magnet makers and traders in Japan/Korea/Europe (generic; specific magnet-maker names not disclosed in sourced material — historically Lynas has not itself made magnets, unlike MP Materials).

Chokepoints / single-source dependencies:

1. LAMP Malaysia is the single separation node for the bulk of output — and it carries political/licensing tail risk (Lens 10). One plant, one host country, historically contested.
2. Kalgoorlie de-risks the cracking step but had power-stability issues during FY2026 that fed through to Malaysian output — an internal single-point dependency.
3. Seadrift, Texas was meant to be the third leg (heavy-REE processing on US soil) but construction was halted over a wastewater-disposal permit, and the original agreement was restructured into the smaller US$96M offtake LOI — i.e. the US processing footprint is, as of mid-2026, in limbo. Prior DoD funding to the project had reached ~US$258M (Aug 2023).

Lens 3 · Competitive Advantages (moats)

The moat is the hardest to replicate asset in the entire critical-materials complex, and it is real:

1. Regulatory + know-how moat (dominant). Separation is a chemically nasty, radioactive-byproduct-generating, permit-intensive process that took Lynas ~12 years and repeated near-bankruptcy to make work at scale. That operating scar tissue is the moat — Iluka's Eneabba refinery won't commission until 2027, Arafura's Nolans is pre-construction awaiting funding, Hastings targets first concentrate (not separation) in Q4 2026. Nobody outside China is close on separated heavy REE — Lynas is a monopoly-of-one there.
2. Owned orebody + integrated flowsheet. Mt Weld gives Lynas captive feedstock at a >20-yr reserve life; MP Materials owns Mountain Pass but is the only comparable integrated Western peer. This is real bargaining power over customers (they need non-China supply more than Lynas needs any single buyer) but weak bargaining power over the price itself — see the fatal caveat below.
3. Government/allied capital as a moat-widener. Japanese (JARE) and US (DoW) capital and price floors partially insulate the P&L and subsidise expansion — a structural advantage emerging peers can't yet access at the same scale.

The fatal caveat on the moat: Lynas controls supply capability but not the price. NdPr clears against a benchmark set by Chinese MIIT quota policy and Chinese spot behaviour; China retains a documented history of flooding the market to undercut Western producers. So the moat protects Lynas's existence and volume, but the earnings are a price-taker's earnings on a commodity a hostile actor can move. That is the central tension of the whole thesis.

Lens 4 · Segments

Lynas does not break out revenue into clean product/geographic segments in the sourced material (the research-layer segments.csv is empty). What can be said with provenance:

• By product: revenue is overwhelmingly NdPr-oxide-driven; La/Ce ("SEG"/lights) are low-value ballast; Dy/Tb (heavies) are new (first sales H1 FY2026) and small by volume but high strategic value and margin. First Dy/Tb sales were cited as a specific NPAT driver in H1 FY2026.
• By geography of processing: mining WA → separation Malaysia → sales predominantly Japan/Asia, with the US offtake newly added.
• Trend: the mix is shifting up the value curve — from lights-heavy toward NdPr + a nascent heavy-REE line — which is the deliberate "Towards 2030" strategy. Segment-level revenue/EBITDA splits: n/a (would require the Appendix 4E segment note, not ingested).

---

Phase B — Measure performance

Lens 5 · Earnings Result (latest print + most recent quarter)

H1 FY2026 (six months to 31 Dec 2025), reported 26 Feb 2026:

• Revenue: A$413.7M, +63% YoY.
• EBITDA: A$152.4M, +300% (from A$38.1M in H1 FY2025); EBITDA margin 36.7% (up from ~15%).
• NPAT: A$80.2M, +1,259% YoY — driven by higher NdPr volumes, sharply higher pricing, and first Dy/Tb sales.
• Cash & short-term deposits: A$1,030.9M (up from A$166.5M at 30 Jun 2025) — the jump is the A$932M equity raise (A$750M institutional placement + A$182M retail SPP) landing, not organic cash generation.
• Production: NdPr 3,407 t (+15% YoY), total REO 6,375 t.
• Avg NdPr selling price H1 FY2026: A$68.4/kg, with the Dec-2025 quarter alone at A$85.6/kg as prices accelerated.

Most recent quarter — Q3 FY2026 (March 2026 quarter Activities Report):

• Revenue A$265M (+115% YoY, +31% QoQ), sales volume 3,131 t (+29% QoQ), avg selling price A$84.6/kg (vs A$50.5/kg in Q3 FY2025), gross margin 42%.
• ROIC cited as capable of lifting toward ~8.5% if Q3 run-rate holds (vs 2.5% in H2 FY2025) — improving but still modest on a >A$1.5B invested base.
• Note the wrinkle: the Dec-2025 quarter saw total REO dip to 2,382 t on kiln maintenance in Malaysia and power instability at Kalgoorlie — so the trend is up and to the right on price but lumpy on volume/execution.

FY2025 (full year to 30 Jun 2025), for context: revenue A$556.5M, profit only A$8M (down from A$84.5M FY2024) — the trough year before the 2026 price surge; management guided FY2026 revenue toward ~A$1.14B. Trailing-twelve-month revenue is A$715.9M, TTM net income A$82.4M.

Read: an operationally leveraged commodity producer at an inflection — margins and profit exploded on price, not primarily on volume. The balance sheet is now fortress-strong (A$1B+ cash, essentially the raise). The quality of the beat is price-driven, which cuts both ways.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on the shelf; sentiment is inferred from sourced call coverage and management public statements.

• Direction of tone: from survival/defensive (FY2025 trough, weak profit, CEO-transition uncertainty) to confident/expansionary through H1–Q3 FY2026. Recurring management themes: "Towards 2030" growth, "first heavy rare earths outside China", disciplined capital, and — pointedly — that "rare-earths buyers must pay more to cut China risks" (Lynas explicitly arguing for a Western risk premium / floor pricing).
• What they started saying: heavy-REE (Dy/Tb/Sm) commercialisation; US government partnership; price-floor advocacy.
• What got quieter: the Seadrift Texas processing build (reframed from a flagship facility to a smaller offtake LOI after the permit halt) — a tell that management is managing down expectations on the US processing leg while keeping the US relationship.
• CEO transition (Lens 9) is the sentiment swing factor: the market took the Lacaze→Le Roux handover as continuity (Le Roux is a 16-year insider), not disruption.

Lens 7 · Comps

Peer table — the only two at-scale integrated Western producers plus the emerging ASX field.

Comps read: there is effectively no clean peer. MP Materials is the closest on strategy but is loss-making (so P/E is meaningless and it trades on option-value/strategic scarcity) and one-fifth Lynas's revenue at comparable market cap — MP carries a bigger strategic premium than Lynas on a per-dollar-of-revenue basis. The rest of the field is pre-revenue or pre-separation — they are call options on catching up, not comparables. Lynas is thus the only investable at-scale profitable Western separated-REE producer, which is exactly why it commands a 200x+ trailing / 37x forward multiple. Whether that multiple is a moat premium or a mania premium is the whole debate (Lens 12/13).

Lens 8 · Stock-Price Catalysts (>5% moves, multi-year pattern)

The tape reveals a name driven almost entirely by geopolitics and the NdPr price, not company-specific earnings execution:

• 2011: all-time high ~A$26.63 (Apr) / A$23.68 (Aug) on the first China-rare-earth-scare spike.
• 2014–2016: catastrophic collapse to an all-time low of A$0.30 (25 Jun 2015) — a ~99% drawdown as China flooded the market and Lynas nearly went bankrupt. This is the single most important fact in the file for sizing risk..
• 2019–2020: Malaysia licensing scares moved the stock sharply (renewal uncertainty = existential for the LAMP-dependent model).
• Apr 2025: China's heavy-REE export-control announcement (7 elements incl. Dy/Tb) — the demand-side catalyst that ignited the 2025 re-rate.
• Jul 2025: MP Materials' DoD deal (US$110/kg floor) re-rated the entire Western REE complex, Lynas included.
• Oct 2025: China's further export-control escalation (extraterritorial "50% rule") — Lynas jumped on scarcity fears.
• Mar 2026: Malaysia 10-yr licence renewal + JARE-to-2038 extension + US DoW LOI — a cluster of de-risking events.
• 2025 into 2026: stock more than tripled, touching a 12-month high ~A$22.37 and a >A$20B market cap, then eased to ~A$18 as the US-China trade temperature cooled.

Pattern: the market reacts to (1) China export/quota policy, (2) the NdPr price, (3) Malaysia licence status, (4) allied-government deals — in that order. It barely trades on a production beat/miss. This is a macro/geopolitical instrument wearing a mining company's clothes.

---

Phase C — Judge people & books

Lens 9 · Management

1. Track record (exceptional, but just handed off). Amanda Lacaze ran Lynas for ~12 years and executed one of the great resource turnarounds: inherited a financially distressed, un-commissioned LAMP under heavy Malaysian community/regulatory attack, and left it as the only at-scale rare-earth processor outside China and the first Western heavy-REE producer. She retired effective end-June 2026.
2. Succession (continuity, not rupture). Pol Le Roux, COO, appointed Interim CEO from 1 July 2026. He joined Lynas in 2010 (16 years) across commercial, downstream and operational roles — a deep insider, not a parachute. Lacaze stays available through Sep 2026; the board is running a permanent-CEO search. Key-person/governance flag: the company just lost its defining operator and has only an interim CEO at the exact moment it is spending >A$1.5B on the biggest expansion in its history. That is a real, if well-managed, transition risk.
3. Capital-allocation history. Reinvestment-led, no dividend. The A$932M equity raise (dilutive but at a strong share price) funds Mt Weld to 12,000 tpa + Malaysian heavy-REE — arguably good use of a high stock price to fund a strategic build. ROIC is improving (2.5% H2 FY25 → possibly ~8.5% if Q3 FY26 holds) but is still low relative to the capital deployed and the multiple.
4. Red flags: none egregious in sourced material — no related-party scandal, no promotional-management pattern beyond normal price-floor advocacy. The equity raise was dilutive but strategically defensible. Executive comp not reviewed (would require the remuneration report — not ingested).
5. Archetype: professional-manager-led (Lacaze and Le Roux are operators, not founders). Appropriate for a capital-intensive processing business at scale — but the founder-grade conviction through the 2015 near-death was Lacaze's, and that is what just walked out the door.

Lens 10 · Forensic Red Flags

Accounting-risk scan (web-only — no filings ingested, so this is directional, not a line-by-line forensic teardown):

• Earnings vs cash quality: the H1 FY2026 cash jump to A$1.03B is financing-driven (the A$932M raise), not operating cash generation — a naïve reader could mistake the balance sheet for organic strength. Flag, not fraud: it's disclosed.
• Price-dependence of earnings: NPAT +1,259% is almost entirely a price effect (NdPr A$50→A$85/kg), not a durable structural margin. High operating leverage means the same mechanism reverses violently if NdPr falls — the P&L has no cushion.
• Capitalised expansion: >A$1.5B invested across Mt Weld/Kalgoorlie/Malaysia; watch for impairment risk on that asset base if NdPr mean-reverts toward the 2023–24 lows (Lynas has a history of the price collapsing). rNPV of the invested base is untested at low prices.
• Inventory/receivables vs revenue, SBC, goodwill, leases: n/a (requires the Appendix 4E / annual report notes, not ingested). Flagged as an open item.

Regulatory findings (required sub-section):

• SEC (EDGAR LR + AAER): regulatory/regulatory-findings.md (generated 2026-07-06) returned 0 findings — because Lynas has no CIK and is not an SEC filer, so no EDGAR search is possible. This is an absence-of-jurisdiction, not a clean-bill-of-health from a US forensic search.
• Non-SEC / Malaysia (the material item): Lynas's LAMP is chronically politically/environmentally contested. Its cracking-&-leaching produces Water Leach Purification (WLP) residue — a thorium-bearing radioactive by-product (half-life ~14bn years). In March 2026 Malaysia renewed the operating licence for 10 years (to 2036, with a 5-year review), but mandated that WLP production cease by 2031 and barred any new permanent radioactive-waste disposal facility. Environmental groups (Greenpeace Malaysia, Friends of the Earth) actively oppose renewal. This is the single largest non-market risk in the file — the entire separation flowsheet depends on a licence in a country where it is politically fraught, and Lynas must re-engineer waste handling before 2031.
• Seadrift/US permitting: construction halted over a wastewater-disposal permit — a concrete regulatory obstacle, not hypothetical.
• Item 3 (Legal Proceedings) from a 10-K: n/a — Lynas files no 10-K.
• Net: No securities-fraud or accounting-enforcement findings surfaced (verified via SEC EDGAR EFTS = no jurisdiction, plus web search as of 2026-07-06). The binding regulatory risk is environmental/licensing in Malaysia (WLP→2031) and permitting in Texas, not accounting.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection (base / bull / bear EPS, next 3 fiscal years)

Lynas's fiscal year ends 30 June; the forward years are FY2026, FY2027, FY2028. FY2026 is nearly booked (H1 done, Q3 done). Every input labeled; outputs ``. The swing variable is the NdPr price, which dominates everything.

Anchors:

• Shares outstanding ~956M.
• FY2026 consensus EPS ~A$0.29; forward P/E 37.5 on A$18.06 → implies forward EPS ~A$0.48 on a next-year basis. The gap reflects which "forward" year the source uses; both are cited.
• H1 FY2026 NPAT A$80.2M actual; Q3 FY2026 revenue A$265M with 42% gross margin.

FY2026 (base): H1 NPAT A$80.2M + a stronger H2 (Q3 already A$265M rev at higher prices) → full-year NPAT ~A$220–260M → EPS ~A$0.23–0.27. Roughly consistent with the A$0.29 consensus.

FY2027:

• Base — NdPr averages ~US$90–100/kg (stabilisation range analysts flag for H2 2026 onward ), volumes rise toward ~9,000–10,000 t NdPr on the Mt Weld ramp → NPAT ~A$300–380M, EPS ~A$0.31–0.40.
• Bull — NdPr holds US$120–135/kg (2026 highs persist on deficit + China controls) → EPS ~A$0.55–0.70.
• Bear — NdPr reverts to US$55–65/kg (China loosens quotas / trade détente / demand air-pocket) → operating leverage reverses → EPS ~A$0.08–0.15, echoing the A$0.08/sh-equivalent thinness of FY2025.

FY2028:

• Base — full 12,000-tpa NdPr run-rate + a nascent heavy-REE contribution + mid-cycle ~US$95/kg → EPS ~A$0.45–0.60. (For reference, third-party narratives model ~A$1.9B revenue / ~A$730M earnings by 2028 — an aggressive bull-ish scenario.)
• Bull — sustained scarcity pricing + heavies scaling → EPS ~A$0.75–1.00.
• Bear — price reversion + Malaysia/Texas execution drag → EPS ~A$0.10–0.20.

The projection is dominated by one exogenous variable Lynas does not control (the NdPr price), which has a demonstrated range of ~US$50 to ~US$135/kg within the last 18 months. Any single-point EPS is false precision; the scenario spread (A$0.10 bear to A$0.90+ bull for FY2028) is the honest answer.

Forecast log: Skipped — per the --watchlist rule (no forecast.ts create in the breadth loop; log a Brier forecast only on a genuinely committed base case in an interactive pass). Candidate base forecast for a future interactive run: "LYC.AX FY2027 NPAT ≥ A$300M, p≈0.50, resolves 2027-08-31."

Lens 12 · Bull vs Bear

Bull case (narrative). Lynas is the physical embodiment of Western supply-chain de-risking — the only profitable, at-scale, separated-REE producer outside China, and the only heavy-REE producer outside China, at the exact moment Beijing has weaponised rare earths (April & October 2025 export controls) and the West has decided at policy level that non-China supply is worth paying a premium for. Demand is secular and inelastic: EV sales ~22.9M units in 2026 (+28% YoY), each NdFeB-motor EV needing ~1–2 kg NdPr, plus wind and robotics — and Western output covers <15% of global demand, a gap analysts say takes 10–15 years to close. Lynas has captive feedstock (Mt Weld, 20+ yr life), a 10-year Malaysian licence, a fortress balance sheet (A$1B+ cash), Japanese offtake locked to 2038 at a US$110/kg floor on 5,000 t/yr, and a US government relationship. It is irreplaceable in a way almost no commodity producer is.

Bear case (2–3 permanent-impairment risks).

1. Price is set by the adversary. China controls ~61% of mining and ~91% of refining and can crater NdPr at will — it did exactly this in 2011–2015, taking Lynas to A$0.30 and the brink of insolvency. The earnings that justify today's multiple exist only at elevated prices China can choose to end.
2. Malaysia single-point dependency + 2031 WLP cliff. The whole flowsheet hinges on one contested plant in one country that has mandated the current waste process end by 2031 — an unresolved re-engineering/relocation problem with a hard deadline.
3. Valuation leaves no margin of safety. 212x trailing / 37x forward P/E, ~2x NAV, P/S ~38x vs a "fair" ~3.9x, DCF ~17% over fair value at ~A$18.19 — the stock already prices a sustained-high-price, flawless-execution world.

Pre-mortem (it's Jan 2028 and the thesis broke — what happened?). A US-China trade détente + a Chinese quota loosening pushed NdPr back below US$60/kg; Lynas's operating leverage reversed and FY2027–28 EPS collapsed to A$0.10–0.15; simultaneously a Kalgoorlie/Malaysia operational stumble or a WLP-related licence tightening dented volumes; the 37x multiple compressed toward 15x on lower, de-rated earnings; the stock round-tripped 40–60% from the ~A$20 highs. The moat held — Lynas still existed and still shipped — but the share price was never about the moat; it was about the price and the multiple.

Are multiples too high? On any conventional metric, yes — this is not priced as a cyclical commodity producer; it is priced as a strategic-scarcity growth asset. That can persist while China controls remain and prices stay high, and it can unwind fast if either breaks.

Contrarian view (what the market refuses to see). Both sides are half-right and the market oscillates between them: the moat is more durable than the bears think (nobody replicates 12 years of separation know-how + a licenced plant + an orebody this decade), but the earnings are far more fragile than the bulls' 37x multiple implies (they are a leveraged bet on a price a hostile state controls). The correct expression is to want to own the asset but to demand a price that respects the A$0.30 tail — i.e. buy the strategic monopoly on a China-détente-driven price washout, not at 2x NAV into a scarcity melt-up.

Lens 13 · Devil's Advocate (short-seller)

Dismantling the bull case:

• What structurally breaks the money-making? A single policy choice in Beijing. China loosening MIIT quotas or dumping inventory resets the NdPr price and, with Lynas's ~37–42% incremental gross margin and high fixed cost (~A$153M/quarter production cost, "largely fixed" ), profit is disproportionately geared to price — a 40% price fall doesn't cut profit 40%, it can eliminate most of it.
• Revenue concentration: heavy reliance on Japan/JARE and now spot pricing; the US processing leg (Seadrift) has already faltered on permits, so the "US pillar" is thinner than the headlines suggest — it's a US$96M offtake LOI, not a built plant.
• Why the moat is weaker than bulls think: it protects volume/existence, not price. Being the only non-China producer is worthless to the share price if China sets NdPr at a level where Lynas earns a low-single-digit ROIC (as it did in FY2025: A$8M profit).
• Most dangerous competitor bulls underestimate: MP Materials — vertically integrating into magnets (Fort Worth "Independence" facility, Apple US$500M, DoD as largest shareholder with a US$110/kg floor on all its NdPr). MP is loss-making today but is building the downstream (magnets) Lynas is not, and has deeper US-government entanglement. If the US "America First" tilt favours the domestic champion, Lynas (foreign, Malaysia-processing) could be disadvantaged in exactly the market it's counting on.
• Worst capital-allocation risk: ploughing >A$1.5B into an asset base whose returns are hostage to a price the company can't set — impairment risk is real if prices revert.
• What must hold for today's price: NdPr sustained ~US$90–130/kg and flawless Mt Weld/Malaysia execution and the Malaysia licence surviving the 2031 WLP transition and the scarcity narrative persisting. That's four things, at least two of which are outside management's control.
• If growth disappoints 20–30%: at 37x forward, a 25% EPS haircut plus a multiple de-rate to ~20x is a ~50%+ drawdown — and the 2015 precedent (−99%) proves the tail is fatter than "50%."
• Single scenario that permanently impairs: Malaysia declines to accommodate the post-2031 waste solution / a future government withdraws the licence — the separation flowsheet loses its node, and years/billions are needed to rebuild it elsewhere. Plausibility: low-to-moderate (the 2026 renewal was granted), but non-zero and existential.

Lens 14 · Management Questions (ordered by information value)

1. What is the concrete, capitalised plan and timeline to comply with Malaysia's 2031 WLP cessation mandate — re-engineer the flowsheet in-country, relocate cracking to Kalgoorlie/elsewhere, or a new waste-neutralisation process — and what is the capex and volume risk of each path?
2. Give the NdPr price at which FY2027 group ROIC falls below your cost of capital — i.e. the price where the >A$1.5B invested base starts destroying value and impairment testing bites.
3. Is the Seadrift, Texas processing facility dead, paused, or alive? If the wastewater permit is unobtainable, what is the actual US processing strategy vs merely a US$96M offtake relationship?
4. With an interim CEO and a permanent search underway, what are the board's non-negotiable criteria for the next CEO, and how do you retain Le Roux and the operating team through the transition?
5. What share of FY2027–28 volume is contracted at floor prices (JARE US$110/kg, DoW) vs exposed to Chinese-set spot, and what is the plan to raise the floored share?
6. How do you compete for US government and OEM favour against MP Materials, given MP's domestic-processing + magnet integration + DoD-shareholder position and the "America First" tilt?
7. Do you intend to integrate downstream into magnet-making (as MP is doing), or remain an oxide producer — and why is that the value-maximising boundary?
8. What is the realistic timeline and cost to reach the 12,000 tpa NdPr run-rate, and what are the gating risks (Kalgoorlie power, Malaysia kiln reliability) after the FY2026 volume wobble?
9. What is the heavy-REE (Dy/Tb/Sm) margin and volume trajectory to FY2028, and could heavies become a material profit line rather than a strategic flag?
10. How do you think about returning capital (buybacks/dividends) vs continued reinvestment, given the balance sheet now holds A$1B+ cash and no dividend is paid?
11. What is your assessment of the structural NdPr supply-demand balance through 2030 — is the current deficit durable, or does Western + Chinese new capacity close it and reset prices?
12. What hedging or offtake structures could reduce the P&L's leverage to Chinese-set spot prices without sacrificing upside?
13. What is the decommissioning/rehabilitation liability at Mt Weld and the Malaysia WLP residue, and how is it provisioned?
14. How exposed is the expansion to reagent, energy, and labour cost inflation in WA, and what's the plan to protect the ~40% gross margin as volumes scale?
15. What would make you walk away from Malaysia entirely, and what is the contingency separation footprint if that day comes?

---