Rio Tinto

Phase A — Understand the business

Company Overview

Rio Tinto is the world's second-largest diversified mining major (behind BHP by market capMarket capitalisationThe share price multiplied by the number of shares. What the market says the equity is worth.) — it digs, processes and ships the bulk industrial and energy-transition metals: iron ore, aluminium (bauxite→alumina→aluminium), copper, and now lithium. FY2025 consolidated sales revenue $57.6bn (2024: $53.7bn, +7%). Market capMarket capitalisationThe share price multiplied by the number of shares. What the market says the equity is worth. ~$162.6bn.

How it actually makes money — one number tells the story: of $25.4bn FY2025 underlying EBITDA, Iron Ore is $15.2bn (60%), Copper $7.4bn (29%), Aluminium & Lithium $4.6bn (18%) — the three sum to more than 100% because "Other/central" is a drag. This is a copper-and-lithium growth story bolted onto an iron-ore annuity. The Pilbara (Western Australia) iron-ore system is the engine; everything else is either a cyclical passenger (aluminium) or a capital-hungry future bet (copper ramp + lithium build).

• Products / commodities:
• Iron ore — Pilbara Blend (60.8% Fe, downgraded from 61.6% in 2025), plus IOC (Canada) pellets/concentrates and the new Simandou (Guinea) high-grade ore that shipped first cargo Dec 2025.
• Copper — Oyu Tolgoi (Mongolia, ramping), Kennecott (Utah), Escondida (Chile, equity stake with BHP), plus gold as a by-product (464koz mined 2025).
• Aluminium — bauxite (Weipa/Amrun/Gove), alumina (Yarwun/QAL), smelted aluminium (Canadian hydro-powered smelters — Kitimat, AP60; plus Boyne, Tiwai Point NZ, ISAL).
• Lithium — the Arcadium platform (Fénix, Olaroz, Sal de Vida, Nemaska, Bessemer City hydroxide) + Rincon (Argentina); targeting 200kt LCE capacity by 2028.
• Customers: China dominates — ~60% of 2025 sales by destination; the balance is Japan, South Korea, other Asia, Europe, US, Australia, Canada. Iron ore is sold to Chinese/Asian steel mills; aluminium to autos/construction/packaging; copper to smelters/wire; lithium to battery/cathode makers.
• Contract structure: Iron ore is largely index-linked — 75% priced on the monthly average index, 10% quarterly-lag, 15% quarterly/other; 25% FOB. So realized price tracks the Platts 62% Fe index almost 1:1 (2025 realized Pilbara $90.0/dmt FOB vs Platts 62% $92.5). This is a price-taker in a commodity market — no take-or-pay pricing power on the flagship product. Aluminium = LME + regional (Midwest/VAP) premiums.
• Suppliers / cost base: diesel, coke, caustic soda, power (huge for aluminium — the smelters' economics are their power contracts), mining contractors, equipment. Grounding: KB commercial-layer files for critical-materials are missing (bottlenecks.md, supply-chain.md, positioning.md all absent per our model) — so Lenses 1–4 are filing- and web-grounded, not commercial-layer-grounded. Flag for coverage backfill.

Strategic frame (new CEO): Simon Trott's "Stronger, Sharper, Simpler" program — $650m annualised productivity benefits already announced (targeting more), ~20% copper-equivalent production growth 2024→2030 (~3% CAGR), and a promise to release $5–10bn in cash from the asset base.

Supply Chain

Map, upstream → Rio → end customer, with named stakeholders:

Iron Ore (Pilbara):

• Upstream: mining fleet + explosives (Orica), autonomous haulage (Caterpillar/Komatsu trucks), diesel, the AutoHaul autonomous rail network.
• Rio nodes: ~16 Pilbara mines → integrated heavy-haul rail (~1,700km) → Dampier & Cape Lambert ports → capesize vessels. Joint ventures: Hamersley, Robe River (with Nippon Steel, Mitsui, Sumitomo Metal Mining as minority JV partners); Hope Downs (50/50 with Hancock Prospecting/Gina Rinehart).
• Chokepoint: the Pilbara is climate-exposed — four Q1 2025 cyclones cost −$0.7bn EBITDA (−$0.6bn volume, −$0.1bn recovery). Port/rail is the throughput constraint; record H2 shipments show the system can flex.
• Downstream: Chinese steel mills (Baowu et al.), plus a portside blending business in China (23.2Mt sold at Chinese ports 2025).

Simandou (Guinea) — the new node: a genuine mega-project supply chain built from scratch — the SimFer mine (Rio + Chinalco-led Chinese consortium + Government of Guinea) + a ~600km trans-Guinean railway + the WCS deep-water port. First train loaded Oct 2025, first shipment Dec 2025, landed China Jan 2026; tertiary crushing done in China. Largest integrated mining+infrastructure project in Africa; could add ~55% to Guinea's GDP by 2030.

Copper:

• Oyu Tolgoi (Mongolia): underground block-cave (development complete Q4 2025) + open pit; conveyor-to-surface now operational. Partner/counterparty risk: Government of Mongolia (34% via Erdenes; new financial terms agreed June 2026 after negotiation; a 2023 export blockade is precedent) — a live sovereign chokepoint. Entrée licence transfer still pending.
• Kennecott (Utah): integrated mine + smelter + refinery (the only fully-integrated US copper smelter) — 45-day planned smelter shutdown in 2025 cut refined volume; geotechnical constraints until higher-grade "Slice 2" access in 2027.
• Escondida (Chile): 30% equity (operator BHP 57.5%, JECO/Japan consortium the rest) — Rio takes dividends, not consolidated cash.
• Nuton: proprietary bioleaching venture — first copper cathode Dec 2025 at Johnson Camp Mine, Arizona; targets primary sulphides (up to 70% of untapped copper) without smelting.

Aluminium & Lithium:

• Bauxite (Weipa/Amrun, Queensland; Gove) → alumina (Yarwun; QAL — 80% Rio / 20% Rusal, but Rusal's tolling is sanctioned, so Rio uses 100% of QAL capacity) → aluminium smelted with hydro power (Canada: Kitimat, AP60 Quebec; NZ: Tiwai Point; Iceland: ISAL). Power is the single-source dependency — NZAS had to curtail on a Meridian Energy call (Mar–Jun 2025).
• Lithium: Argentina brine (Fénix, Olaroz, Rincon, Sal de Vida) + Canada (Nemaska hard-rock) + US (Bessemer City hydroxide conversion). Mt Cattlin spodumene (WA) placed on care & maintenance Mar 2025; Jadar (Serbia) placed on care & maintenance Nov 2025.

Verdict on the chain: world-class, vertically integrated, but with three named sovereign/counterparty chokepoints — Guinea (Simandou + a live AFP bribery probe), Mongolia (Oyu Tolgoi terms), and China (60% of revenue + JV partner in Simandou). Names present ✓.

Competitive Advantages (moats)

• Cost-curve position (the real moat): Pilbara iron ore at $23.5/t unit cost against a $90/dmt realized price = a >60% cash margin and 39% segment ROCE even in a down-price year. Rio sits in the first quartile of the global iron-ore cost curve alongside BHP and Vale — the ~370Mt of marginal, high-cost "junior" supply is the price floor, and Rio prints cash at prices that bankrupt the marginal tonne. This is a durable, scale-and-geology moat; you cannot replicate the Pilbara.
• Irreplaceable orebodies + infrastructure: the Pilbara rail/port system, Oyu Tolgoi's block-cave (one of the largest known copper-gold deposits), Simandou's high-grade (65% Fe) deposit — decades of reserves behind 100-year-old franchises. Regulatory + capital + time barriers make these effectively un-newbuildable.
• Project-delivery capability: Simandou delivered first ore <2 years after major construction began — management explicitly frames "partnership as a Rio Tinto superpower." Credible; it's the differentiator vs. peers who blow up mega-project budgets.
• Diversification (2025's proof point): a 6% lower iron-ore index price was fully offset by higher bauxite/alumina/aluminium/copper/gold — group EBITDA still rose 9%. The portfolio is the shock absorber.
• Bargaining power — asymmetric: Weak on iron ore (index price-taker into a China-concentrated buyer base — the buyers need the tonnes but set the price via the index). Stronger on copper/lithium where supply is genuinely scarce. Over suppliers (contractors, equipment) Rio has scale leverage.
• Balance sheet as a moat: single-A credit rating, disciplined capital framework — lets Rio counter-cyclically acquire (Arcadium at the lithium trough) and out-invest weaker miners.

Moat gaps: no product differentiation on the flagship (iron ore is a graded commodity, and Rio downgraded its blend to 60.8% Fe in 2025 — a quality erosion). Lithium is a new business with no moat yet (commodity brine/hard-rock; Rio is a price-taker there too, buying in near the bottom). ESG/social-license is a recurring moat liability (Juukan Gorge, Bougainville, Guinea) rather than an asset.

Segments

FY2025 by product group [all research-layer: filings/20-f-2025-q4.md]:

(Segmental revenues sum to ~$59.8bn > $57.6bn consolidated because of intersegment/portside/freight reconciliation and Simandou being reported outside segments during ramp.)

By geography (revenue by destination): Greater China ~60%, then US, Japan, Europe, other Asia, South Korea, Canada, Australia. This is the single most important risk fact in the whole dossier — 60% of the book rides on China, and within that, on Chinese steel demand (iron ore) which is in structural plateau.

Trend & cause:

• Iron Ore — decelerating: revenue −8%, EBITDA −11%, ROCE 48%→39%. Cause: realized price $97.4→$90.0/dmt (−8%), a −$2.3bn price hit, partly offset by +1% shipments and lower SP10 mix. Volume flat (327Mt) against cyclones. This is the mature-annuity-in-slow-decline chart.
• Copper — accelerating hard: EBITDA +114%. Cause: Oyu Tolgoi copper +61%, gold +65%, realized copper 422→457c/lb, C1 unit cost crashed 142→67c/lb. FCF +437% to $2.8bn. This is the growth engine finally paying off.
• Aluminium — cyclically strong: EBITDA +29% on realized price $2,834→$3,318/t (LME +9% + premiums), record bauxite (62.4Mt) — but ~$1bn eaten by US Section 232 tariffs (lost the 10% exemption Mar 2025, now 50%; Midwest premium has since compensated).
• Lithium — embryonic: only $0.2bn EBITDA, LCE 57kt (46kt attributable to Rio), Arcadium consolidated from March. Consuming $1.5bn cash. The 200kt-by-2028 target is the whole thesis; today it's a rounding error.

---

Phase B — Measure performance

Earnings Result (FY2025 annual print)

Rio reports semi-annually, so the "latest print" is the full-year FY2025 result (20-F, 2026-02-19). All:

What drove it: underlying EBITDA +$2.0bn — volumes & mix +$2.4bn (copper +61% at Oyu Tolgoi, record bauxite), price movements net ~neutral (iron ore −6% offset by bauxite/alumina/aluminium/copper/gold), cost discipline −5% real unit costs, partly offset by cyclones (−$0.7bn), inflation, and Simandou ramp opex.

Why net earnings fell 14% while underlying was flat: (1) higher D&A (Oyu Tolgoi ramp + Arcadium) −$0.6bn; (2) tax rate jumped 28.3%→31.5% (Escondida mix, unrecognised DTAs) −$1.0bn; (3) higher finance costs on the +$8.9bn debt; (4) −$1.6bn of items excluded from underlying (impairments/one-offs). Statutory profit is the honest number and it's down.

Balance-sheet flags: the marquee event is net debt +$8.9bn to $14.4bn, driven by the $7.6bn Arcadium cash acquisition + $6.1bn dividends paid, funded by $9bn of new bond issuance, partly offset by $4.0bn FCF + $1.0bn NCI project funding. FCF halved-ish (−28%) because capex ran to $11.4bn (Simandou + Pilbara replacement mines + lithium). Still single-A; management runs "principles-based, no net-debt target."

Guidance/outlook & tone: 2026 copper guidance 800–870kt (Oyu Tolgoi ramp); Pilbara mid-term capacity 345–360Mtpa; lithium 200kt LCE by 2028; ~20% CuEq growth 2024→2030; $5–10bn cash-release target. Trott's tone is operational-discipline-and-deliver, a deliberate shift from Stausholm's rebuild-trust register.

Unusual vs. own history: the debt step-change and the FCF compression are both out of character for post-2016 Rio, which had been a fortress-balance-sheet, buyback-and-special-dividend machine. This is a deliberate pivot from returning cash to deploying it — the single most important behavioural change in the file. (No special dividend since 2022; ordinary dividend flat.)

Earnings Calls (sentiment trend)

No transcripts/ on the shelf (transcripts=0) — sentiment is read from the 20-F narrative + CFO/CEO statements +. Rio's half-yearly cadence means fewer, denser communications than a US quarterly filer.

Tone arc (Stausholm era → Trott era):

• 2021–2024 (Stausholm): register = "rebuild trust, decarbonise, be a better partner" post-Juukan Gorge. Heavy ESG/social-license language. Consistent-returns messaging.
• 2025 (Trott, from Aug): register hard-pivots to "Stronger, Sharper, Simpler," productivity, unit costs, project delivery, release cash." The CFO statement leads with "leaner… strong focus on productivity and efficiencies." New phrases: "$650m productivity benefits," "release $5–10bn," "copper equivalent growth," "capital intensity."
• Things they stopped emphasising: the reflexive ESG-first framing softened; special dividends absent; the word "trust" recedes in favour of "performance."

Recurring focus items: safety (two SimFer/Guinea fatalities weigh on the narrative and cut the 2025 STIP — this is genuine, not boilerplate), Safe Production System (now at all sites), copper/lithium growth delivery, cost discipline. The sentiment shift is from custodianship to operating aggression — a new CEO clearing the decks. Constructive if delivered; a red flag if "sharper" becomes short-termism on safety or maintenance.

Comps

Peer set = the diversified/iron-ore majors. Multiples are, dated; where not sourced, "n/a." As-of ~2026-07-06 unless noted.

(5-yr avg ROE not separately sourced per name — flagged n/a rather than fabricated. RIO's own ROCE is 16% underlying / ROE 16.4%.)

Read: RIO trades at a premium to the pure iron-ore names on EV/EBITDA (8.95x vs Fortescue 5.55x, Vale 5.83x) and roughly in line-to-slightly-cheap vs BHP (7.38x) — but BHP earns a materially higher ROE (24.7% vs 16.4%) and carries less iron-ore concentration risk, so BHP looks like the higher-quality diversified major on this snapshot. The market is paying up for RIO's copper+lithium growth optionality (the forward P/E compresses to 11.6x on expected earnings growth), while punishing Vale (governance/Brazil/Samarco overhang → cheap EV/EBITDA + fat 5.9% yield) and treating Fortescue as a high-beta iron-ore-only trade. RIO sits in the middle: better diversified than Fortescue/Vale, lower-returning and more iron-ore-levered than BHP. Fair, not cheap.

Stock-Price Catalysts (>5% moves, last ~5yr)

Mostly; the pattern matters more than any single date.

• 2021 super-cycle peak: iron ore >$200/dmt on China stimulus → record FY2021 (revenue $63.5bn, PAT $22.6bn, EPS 1,304.7c, dividends incl. specials ~$16bn). The stock and payout peaked here. Everything since is the normalization off that peak.
• 2022–2024 grind lower: iron ore mean-reverted to ~$100/dmt; revenue slid $63.5→$53.7bn; the shares de-rated with the price and the fading China property cycle. China stimulus headlines = the dominant swing factor.
• 2020 Juukan Gorge: destruction of a 46,000-year-old Aboriginal site → CEO + two execs out, a permanent social-license scar and a reputational reset. Governance/ESG shocks move this stock.
• Oct 2024 – Mar 2025 Arcadium: announced ~$6.7bn (headline) / $7.6bn cash lithium acquisition — market reception broadly positive as counter-cyclical bottom-buying; completed Mar 2025.
• 2025 recovery: RIO shares +~34% in 2025 on surging copper + aluminium prices and Arcadium optimism; 52-wk range $57.66–$112.58, now ~$93.77.
• Dec 2025 Simandou first shipment: de-risking milestone for the biggest project in the portfolio.
• May/Aug 2025 CEO change: Stausholm → Trott; a strategy-register catalyst (Stronger/Sharper/Simpler).
• June 2026 Mongolia terms: new Oyu Tolgoi financial terms agreed with the government — resolves an overhang.

What the market actually reacts to (pattern): #1 the iron-ore price / China steel-demand signal (still the single biggest driver despite diversification); #2 mega-project + M&A milestones (Oyu Tolgoi, Simandou, Arcadium); #3 governance/ESG/sovereign shocks (Juukan, Guinea, Mongolia). It is becoming a copper story but is still traded as an iron-ore/China proxy.

---

Phase C — Judge people & books

Management

• CEO — Simon Trott (from Aug 2025). Rio veteran (decades): former Chief Executive, Iron Ore and before that Chief Commercial Officer — i.e. he ran the profit engine and sold the product. Track record: credible operational + commercial delivery in the Pilbara; his signature is the "Stronger, Sharper, Simpler" reorg (collapse to 3 product groups, $650m productivity, cash-release). Archetype: professional operator/insider, not a founder or a dealmaker — appropriate for a mature major that needs execution and cost discipline, less obviously the visionary you'd want steering a $7.6bn lithium bet into a new industry. Too new to grade on capital allocation.
• Predecessor — Jakob Stausholm (CEO 2021–2025, ex-CFO from 2018). Under him: returned >$40bn to shareholders, rebuilt reputation post-Juukan, and advanced Simandou + Oyu Tolgoi to the finish line, and pulled the trigger on Arcadium. A strong, disciplined-steward tenure; his exit was orderly (not a scandal).
• Chair — Dominic Barton (ex-McKinsey global Managing Partner, ex-Canada ambassador to China). Heavyweight board leadership with deep China relationships — relevant given 60% China revenue. Signed the FY2025 strategic report.
• CFO — Peter Cunningham. Signed the CFO statement; steward of the single-A / no-net-debt-target framework.
• Tenure & skin in the game: deep-bench insiders (Trott decades at Rio); insider ownership modest in $ terms as is normal for a mega-cap plc — no our figures on the shelf, so unquantified. Big reorg = several senior departures in 2025 (Sinead Kaufman/Minerals, Kellie Parker/Australia both stepped down; Matthew Holcz → CEO Iron Ore).
• Capital-allocation history: the defining trait. A 60% payout for 10 straight years (top of the 40–60% policy) — a genuine dividend aristocrat of mining. Historically fortress balance sheet + specials at the peak. FY2025 = a deliberate pivot to deployment: Arcadium ($7.6bn), Simandou build-out, Pilbara replacement mines — funded by debt, at the cost of FCF and specials. The bet: counter-cyclical lithium + copper growth beats returning cash. ROCE fell to 16% on the higher asset base — the market will judge them on whether that reverses as Oyu Tolgoi/Simandou/Rincon ramp. Legacy caution: the 2007 Alcan acquisition ($38bn) is the cautionary tale — Rio overpaid at the top and carried the debt for years; the risk is Arcadium/lithium becomes "Alcan 2.0" if lithium stays depressed.
• Red flags: (1) safety — two fatalities at SimFer/Guinea (Mohamed Camara + a contractor, Feb 2026) cut the 2025 STIP and are the sharpest cultural blemish; (2) the live Australian Federal Police investigation into 2011 Guinea contractual payments (see Lens 10) is an unresolved integrity overhang; (3) the "Sharper/Simpler" cost push must not become deferred maintenance. No promotional/related-party comp red flags surfaced.

Forensic Red Flags

Acting as a forensic analyst on the FY2025 20-F + regulatory findings.

• Revenue recognition / provisional pricing: iron ore and copper carry mark-to-market provisional pricing — copper's provisional adjustment added +$758m to 2025 revenue (2024: −$92m). This is legitimate IFRS but flatters the YoY copper revenue optics — strip it and copper's underlying beat is smaller. Watch this line; a swing to negative provisional pricing would reverse a chunk of the copper story.
• Underlying vs. statutory gap: underlying earnings $10.9bn vs net earnings $10.0bn — a −$1.6bn wedge of "items excluded from underlying" (impairments, one-offs). The gap is modest (net is ~92% of underlying) and the direction is honest (statutory is lower), so the non-GAAP is not egregiously flattering — but the headline the company leads with (669c underlying EPS, "flat") masks a −14% statutory decline. Read the statutory number.
• Cash flow vs. earnings: net cash from ops $16.8bn comfortably exceeds net earnings $10.0bn (heavy D&A on a capital-intensive base) — cash conversion is healthy, no earnings-quality red flag there. But FCF −28% to $4.0bn while net debt +$8.9bn is the tension: growth capex + Arcadium + dividends now exceed self-funding. Not a fraud flag; a capital-cycle flag.
• Goodwill/intangibles: the $7.6bn Arcadium deal loads the balance sheet with lithium goodwill/intangibles into a depressed lithium price — impairment risk is real if lithium doesn't recover (total assets jumped $102.8bn→$128.1bn largely on this). Jadar already parked (care & maintenance Nov 2025) and Mt Cattlin idled — early signs the lithium portfolio is being rationalised. Watch for a lithium write-down at H1 2026.
• Tax: effective rate on underlying rose to 31.5% (from 28.3%) — flagged by management (Escondida mix, unrecognised DTAs, prior-year adjustments). Not manipulation; a real earnings headwind.
• SBC / related parties: no unusual stock-comp or related-party flags in the disclosure. Contingent liabilities $322m (2024: $192m) — small relative to scale.
• Closure/restoration provisions: note that Canadian aluminium smelters carry no closure provision because "the date of closure cannot be reliably estimated" (indefinite-lived hydro assets). Defensible under IFRS but means a category of long-tail environmental obligation sits off the recognised balance sheet.

Regulatory findings (required):

• SEC (EDGAR EFTS): 0 Litigation Releases and 0 AAERs naming Rio Tinto in the 2021-07-06→2026-07-06 window.
• 2011 Guinea/Simandou contractual payments (Item / contingencies): Rio resolved a self-disclosed SEC investigation in 2023 into US$10.5m paid to a consultant in 2011 re Simandou; the UK SFO closed its case Aug 2023; but the Australian Federal Police maintains a live investigation — "the outcome remains uncertain, but it could ultimately expose the Group to material financial cost. No provision has been recognised.". This is the single named open legal exposure — quote directly.
• Bougainville / Panguna (PNG): a 2024 class action in PNG's National Court against Rio + BCL was dismissed entirely in Sept 2025; an appeal is filed in the PNG Supreme Court. Rio "will strongly defend." Legacy-impact remediation (PMLIA) ongoing.
• Juukan Gorge (2020): ongoing modernisation of Traditional Owner agreements; provisions created for historical claims; "process is incomplete… further claims could arise."
• Rusal / QAL sanctions: Australian govt sanctions prevent QAL tolling for 20%-owner Rusal; Rio uses 100% of capacity — a geopolitical entanglement, not an enforcement finding.
• Non-SEC web scan ("Rio Tinto" (FTC OR DOJ OR FDA OR settlement OR fine OR penalty) enforcement): no new material enforcement hit beyond the Guinea/AFP matter surfaced in this pass; the Guinea probe is the live one.

Net Lens-10 verdict: accounting quality is solid (statutory below underlying, strong cash conversion, no SEC/AAER findings, small contingencies). The genuine risks are (1) lithium goodwill impairment into a weak price, (2) the copper provisional-pricing tailwind that could reverse, and (3) the unresolved AFP Guinea bribery probe. Not a forensic short; a "watch the write-downs and the Guinea headline" file.

---

Phase D — Project & stress-test

Forward Projection (EPS, next 3 fiscal years)

Base = FY2025 underlying EPS 669.2c ($6.69) on ~1,624m shares. FY ends 31 Dec, so the three years are FY2026 / FY2027 / FY2028. Inputs labeled; outputs ``.

Key drivers:

• Iron ore (60% of EBITDA): 2026 price consensus ~$94–102/dmt CFR (Fitch $100, Deutsche $102/BMI $95, overall ~$94) — roughly flat-to-down vs 2025's $90 realized FOB; volume flat-to-up as Pilbara pushes toward 345–360Mtpa and Simandou adds ~15–20Mt (Rio share) into ramp. Net: iron-ore EBITDA flat-to-down (price soft, Simandou is dilutiveDilutionIssuing new shares, so each existing share owns a smaller slice of the same company. to the market even if accretive to Rio).
• Copper (the growth): Oyu Tolgoi to ~500ktpa (100%) by 2028; 2026 guidance 800–870kt group. Copper price assumed stable-to-firm on electrification demand. Copper EBITDA the primary grower — could add $2–4bn by 2028 vs 2025's $7.4bn.
• Aluminium: cyclical; tariff drag (~$1bn) partly offset by Midwest premium; assume flat-to-modestly-up.
• Lithium: 200kt LCE by 2028, but price-dependent; EBITDA scales from ~$0.2bn toward $0.5–1.0bn only if lithium recovers — kept conservative.
• Costs/leverage: $650m+ productivity, unit-cost discipline; offset by Simandou opex during ramp and higher D&A/finance on the debt.
• Tax: ~31–32% underlying rate assumed to persist. Shares: ~flat (no buyback with debt elevated; dividend absorbs FCF).

Scenario EPS:

Cross-check vs. market: consensus fwd EPS proxies ~$8.64 (Simply Wall St, 2026) sit at the top of my base / into my bull — implying the street is pricing the copper+lithium ramp landing cleanly. On $93.77, fwd P/E 11.6x ⇒ implied forward EPS ~$8.1 — so the market is roughly at my optimistic base. My base ($6.80→$8.00 over 3yr) is a touch below consensus — i.e. consensus may be modestly optimistic on the ramp timing / iron-ore hold.

No our model create in this unattended watchlist pass (per SKILL.md — only log a Brier forecast on genuine commitment). If promoted to a thesis, log: "RIO FY2027 non-GAAP/underlying EPS ≥ $7.20, p≈0.50, resolves 2027-12-31."

Bull vs Bear

Bull case. Rio is a first-quartile-cost, irreplaceable-asset cash machine that is quietly transforming from a China-iron-ore proxy into the West's premier energy-transition metals major. The iron-ore annuity (39% ROCE at a low price) funds everything. Three growth legs are now inflecting at once: Oyu Tolgoi copper +61% and heading to 500ktpa (world-top-5 copper mine), Simandou first ore shipped (a second high-grade iron-ore franchise + African beachhead), and a counter-cyclically-acquired lithium platform (Arcadium at the trough → 200kt by 2028). C1 copper cost crashed to 67c/lb. A 10-year 60%-payout record + single-A balance sheet + Dominic Barton's China access. If copper stays scarce and lithium recovers, the mix re-rates the multiple away from "iron-ore cyclical" toward "structural-scarcity compounder." The market pays 11.6x forward for that optionality — cheap if the ramps land.

Bear case (permanent-impairment risks).

1. China steel demand structurally rolls over. 60% of revenue, and iron ore is 60% of EBITDA — a China property/steel down-leg (the property sector is already contracting; only infrastructure/export offset it) takes iron ore below $85 and craters the engine that funds the whole growth pivot. This is the existential one.
2. Simandou is self-inflicted supply. Rio is adding high-grade, low-cost tonnes into a plateauing market — Simandou (Rio's + the Chinese consortium's + Guinea's) plus other new supply could push the iron-ore price down, cannibalising the Pilbara annuity to grow volume. Growth that destroys price is value-destructive.
3. Lithium = Alcan 2.0 risk. $7.6bn deployed into a commodity at/near a cyclical bottom, with no moat; Jadar already parked, Mt Cattlin idled. If lithium stays depressed, expect goodwill impairments and a "we overpaid at the top of electrification hype" narrative — echoing the 2007 Alcan mistake.

Pre-mortem (18 months out, thesis broke): iron ore fell to ~$80 on a China steel contraction; Simandou's ramped tonnes made it worse; a lithium write-down hit H1 2026 results; Oyu Tolgoi slipped on Mongolia licence friction; FCF stayed compressed so the dividend got trimmed off its 60% top-of-range — and the "diversification re-rating" never came because the market re-anchored on the still-dominant iron-ore exposure. The stock de-rated back toward the low end of its range.

Are multiples too high? No — 8.95x EV/EBITDA / 11.6x fwd P/E is reasonable-to-full for a first-quartile major, below the diversified-major top of range but above pure iron-ore peers. You are paying a modest premium for growth optionality that is mostly still promise. Not expensive; not a bargain.

Contrarian view (what the market refuses to see): The bull consensus treats Simandou as an unalloyed positive. The contrarian read is that Simandou is bearish for Rio's own iron-ore economics — it converts a scarcity-priced annuity into a volume game just as demand plateaus, and the real prize (copper/lithium) is still years and a lithium-price recovery away from mattering. Meanwhile the market keeps trading RIO as a China proxy, so the diversification it's paying for hasn't actually de-risked the tape. The mispricing cuts both ways: bears underrate the copper cash-flow inflection; bulls overrate how quickly the mix stops being iron-ore-and-China.

Devil's Advocate (short-seller)

Dismantling the bull case:

• What structurally breaks it: a sustained China steel contraction. Full stop. 60% of revenue destination, 60% of EBITDA in iron ore, index-linked price-taking — there is no pricing power to defend the flagship if Chinese demand cracks. Diversification is real but insufficient: copper+lithium+aluminium can't backfill a $5–8bn iron-ore EBITDA hole for years.
• Revenue concentration: Greater China ~60%. A geopolitical rupture (Australia-China tension recurrence, tariffs, a forced supply-chain decoupling) or a demand collapse hits harder than any peer except Fortescue. And within Simandou, the customer, the JV partner, and the destination are all China — triple concentration on one counterparty.
• Why the moat is weaker than bulls think: the iron-ore moat is cost-curve, not pricing — it protects margin, not price. And Rio just downgraded its Pilbara Blend to 60.8% Fe — a quality erosion that narrows the premium. Lithium has no moat at all (commodity brine/rock). Copper's moat is real but the growth is capital-hungry and sovereign-exposed (Mongolia).
• Most dangerous competitor bulls underestimate: BHP — higher ROE (24.7% vs 16.4%), less iron-ore-concentrated, a cleaner copper-growth story (Escondida operator + Copper South Australia), and it didn't bet $7.6bn on lithium at the top. On this snapshot BHP is the better-quality vehicle for the same secular thesis. Also Vale on cost (cheap EV/EBITDA) and the Chinese majors + junior supply that set the marginal iron-ore price.
• Worst capital-allocation moves: history says Alcan (2007, $38bn, top-tick, years of debt). The live analogue is Arcadium ($7.6bn into depressed lithium) — plausibly the next value-destroyer if lithium doesn't recover. Deploying into growth and holding a 60% payout is being funded by a +$8.9bn debt swing — you can't do both forever without either the ramps paying off or the dividend giving way.
• Assumptions that must hold for today's price (~$93.77, 11.6x fwd): (1) iron ore holds ~$95+; (2) Oyu Tolgoi hits ~500ktpa on schedule; (3) lithium recovers enough to avoid big impairments; (4) China steel doesn't roll over; (5) copper's +$758m provisional-pricing tailwind doesn't reverse.
• If growth disappoints 20–30%: an iron-ore price to ~$75 + a slipped copper ramp takes underlying EPS toward my bear ($5.50–5.80) → at an unchanged 11.6x that's a ~$65–67 stock (−30%), back to the bottom of the 52-week range ($57.66 low), before any dividend cut or lithium write-down amplifies it.
• Single permanent-impairment scenario & plausibility: a multi-year China steel structural decline (property deleveraging + peak-steel demographics) that resets iron ore to a $60–70 regime. Plausibility: moderate-and-rising — it's the single biggest, most-discussed risk in the sector, and Rio is more exposed to it than BHP. It wouldn't bankrupt Rio (first-quartile cost survives), but it would permanently impair the earnings power that funds the growth pivot and the dividend.

Management Questions (ordered by information value)

1. Simandou: at what iron-ore price does Simandou's added supply become net negative to group iron-ore EBITDA (cannibalising Pilbara price faster than it adds volume)? Show the break-even.
2. China: what is your internal base case for Chinese crude-steel demand 2026–2030, and how much of group EBITDA survives a 15% structural decline in Chinese steel?
3. Lithium/Arcadium: what lithium price is embedded in the Arcadium carrying value, and at what price do you take a goodwill impairment? Is 200kt-by-2028 still economic at spot?
4. Capital framework: with net debt at $14.4bn and FCF compressed, which gives first if commodities soften — the 60% payout, the growth capex, or the single-A rating?
5. Copper provisional pricing: how much of 2025 copper revenue was the +$758m provisional tailwind, and what's the sensitivity if it swings negative in 2026?
6. Oyu Tolgoi: what are the specific gating items (Entrée licence transfer, Mongolia terms, Panel sequencing) to the 500ktpa 2028 target, and the probability-weighted timeline?
7. ROCE: you're at 16% underlying, down 2pp. What's the path back above 18% as the growth assets ramp, and by when?
8. Cash release: the $5–10bn "release cash from the asset base" — which specific assets (Borates? Iron & Titanium? Others) and over what timeframe?
9. Safety: after two SimFer fatalities, what concretely changes in the Guinea/Simandou operating model, and how do you reconcile "Sharper/Simpler" cost pressure with safety investment?
10. Guinea AFP probe: what is the realistic range of financial exposure from the live Australian Federal Police investigation into the 2011 payments, and why no provision?
11. Aluminium tariffs: is the ~$1bn Section 232 tariff drag structural, and does it change your US aluminium footprint/strategy?
12. Pilbara quality: the blend dropped to 60.8% Fe — what's the multi-year iron-content trajectory and the realized-price impact of grade decline?
13. M&A discipline: given Alcan and now Arcadium, what governance guardrails prevent another top-of-cycle mega-deal?
14. Decarbonisation: the 2030 50%-emissions-cut requires timely Pacific Aluminium renewable deals — what's the probability you miss the 2030 target, and the cost of the $8.5bn private renewable underwrite?
15. Trott's mandate: 18 months in, what is the one metric by which shareholders should judge whether "Stronger, Sharper, Simpler" worked?

---