FMG: Iron Ore Giant - China's Property Crisis Casts Long Shadow

Fortescue Metals Group is Australia's #4 iron ore producer (12.4% global seaborne share) trading 21% above fair value at $22.32 AUD versus $18.42 USD equivalent, with asymmetric downside risk from China property sector exposure (>90% sales) offsetting near-term margin recovery potential as Platts iron ore prices normalise and Iron Bridge magnetite ramps production.

• Market Position: Cost-competitive challenger (#2 globally at $18.0/wmt C1 cost) with infrastructure barriers (Port Hedland duopoly) providing 5-7 year moat, but competitive advantage fading predictably as resource depletion and peer automation adoption close cost gap by 2030.
• Financial Performance: EBITDA margins recovering from 54.1% trough (FY25) toward 59.8% (FY28) driven by Platts recovery $102→$115/dmt and Iron Bridge ramp 7→19Mt, generating 14% expected annual returns over 3 years despite volume ceiling at 210Mt infrastructure capacity.
• Valuation: Dynamic weighted fair value $18.42 USD (DCF 38.8%, Multiples 33.4%, NAV 25.5%) implies 21% overvaluation at current $22.32 AUD ($14.49 USD equivalent), with 90% confidence interval $15.66-$21.18 reflecting low 48% terminal value dependency and high 86.6/100 reliability score.
• Investment Assessment: Probability-weighted scenarios (Base 60% $19.45, Bear 30% $12.50, Severe 10% $7.81) produce 40% combined downside probability, suitable for commodity-exposed investors with 3-5 year horizon and tolerance for China concentration risk, but not compelling given overvaluation and fading post-2030 advantages.

Primary catalyst risk: China property crisis deepening (construction -30% YoY) could trigger Bear case within 12-18 months, whilst Iron Bridge achieving 22Mt nameplate ahead of FY28 schedule represents key upside validation point.

Fortescue Metals Group operates as Australia's fourth-largest iron ore producer, extracting and exporting 198Mt annually from Pilbara hematite deposits (58% Fe grade, 172Mt) and Iron Bridge magnetite operations (67% Fe grade, 7Mt ramping to 19Mt by FY28). The business model centres on cost leadership (#2 globally at $18.0/wmt C1 cost versus peer average $20) combined with infrastructure advantages—Port Hedland duopoly berth access and freight proximity to Chinese steel mills ($10-15/dmt advantage versus Brazilian competitor Vale). Revenue generation follows commodity price-taker dynamics, with 96% derived from iron ore sales realising 83-87% of Platts 62% index (grade discount) for hematite and 114% premium for magnetite. Founder Dr. Andrew Forrest's 36% ownership via Tattarang provides strategic continuity across 21-year tenure, driving $6bn Real Zero 2030 decarbonisation commitment (190MW solar, 60% autonomous fleet, 95ML diesel reduction), though Energy division commercialisation failures (-$2.0bn losses FY22-25, Arizona/PEM50 cancelled post-FID) bifurcate execution credibility (Metals 8.5/10, Energy <5/10).

FY25 financial performance reflects late-cycle commodity trough positioning, with revenue declining 15% to $15.5bn (Platts 62% index $119→$102/dmt = 14% price compression) despite volume growth +4% (179Mt shipments versus 173Mt prior year, driven by Iron Bridge initial ramp). EBITDA margins compressed to 54.1% from historical 60% average as operating deleverage (fixed costs $1.7bn persistent despite revenue decline) and hematite realization deterioration (87%→83% = product mix shift toward lower-grade ore) offset cost discipline achievements (C1 cost $18.24→$17.99/wmt = first decline since FY20, validating $500m automation investment). Balance sheet strength remains exceptional with net debt $1.1bn (0.1x EBITDA), $4.3bn cash, and 23x interest coverage enabling through-cycle dividend maintenance (77.5% payout ratio, $1.10 DPS) and counter-cyclical capital deployment. Free cash flow generation $3.1bn (estimated, 46% EBITDA conversion) sustained despite elevated capex $4.4bn (decarbonisation $1.0bn, Iron Bridge completion $0.5bn, sustaining $2.3bn).

Current competitive positioning reflects #4 global scale (12.4% seaborne market share, 198Mt production versus Rio Tinto 330Mt, Vale 310Mt, BHP 250Mt) with #2 cost advantage ($18.0/wmt trailing only Rio Tinto $16.5/wmt) providing 5-7 year Competitive Advantage Period. Infrastructure barriers include Port Hedland duopoly (40% FMG berth allocation, $15bn replacement cost, 8-10 year approval timeline for competing Anketell port) and Pilbara geology advantages (low 1.6x strip ratio, shallow <150m deposits). However, moat fading predictably as resource depletion forces higher strip ratios (1.6x→1.9x by 2035 at new hubs Nyidinghu/Mindy South, increasing C1 cost $18→$20-21/wmt) and peer automation adoption (BHP/Rio targeting 70% autonomous fleets by 2030 versus FMG's current 60%) closes technology gap. Strategic initiatives centre on Iron Bridge magnetite ramp (7Mt FY25 → 19Mt FY28 base case, 67% Fe premium product realising 114% Platts index) and Real Zero decarbonisation (first operational solar 190MW Q1 FY26, 10 electric excavators deployed, targeting $80-100m/yr diesel savings by FY28), though commercial payoff uncertain if green iron premium fails to materialise.

Value creation over the next 12-24 months hinges on three sequential catalysts with distinct probability-weighted outcomes. Primary driver involves Platts iron ore price mean reversion from current $102/dmt trough toward $115/dmt mid-cycle (5-year average), representing 13% recovery that mechanically translates to $2.8bn EBITDA expansion via 90% commodity price pass-through. Near-term validation centres on Q2 FY26 production update (January 31, 2025) where Iron Bridge achieving >4Mt quarterly run-rate (annualised 16-20Mt) confirms base case trajectory, whilst sustained Platts 3-month average >$108/dmt signals mean reversion thesis on track. Operating leverage amplification (DOL 2.4x) should drive EBITDA margin recovery from 54.1% trough (FY25) toward 59.8% (FY28) as fixed costs $1.7bn spread over +$4.5bn incremental revenue, though execution dependent on Iron Bridge air classification technology achieving >88% yield at scale (historical delays FY25→FY28 nameplate create 30% probability Bear case ceiling at 15-16Mt).

Secondary catalyst involves China demand trajectory, where property sector stabilisation (construction starts moderating from -30% YoY decline toward -15-20%) would validate base case steel production plateau at 1.01bn tonnes versus Bear case structural decline to 0.95-0.96bn tonnes. Medium-term monitoring focuses on China Steel PMI remaining >48 (expansion threshold) for consecutive months and PBOC stimulus package scale (>$1trn fiscal support required to offset property wealth effect destruction). Competitive dynamics evolution centres on peer automation adoption timeline—if BHP/Rio delay 70% autonomous fleet targets beyond 2030 (versus current 2028-30 guidance), Fortescue's cost advantage duration extends 2-3 years (CAP 7-9 years versus base 5-7 years), adding $1.50-2.00/share terminal value. Conversely, Simandou (Rio Tinto, Guinea) accelerated first ore to 2028 (versus 2030+ expected) would flood market with 110Mt/yr high-grade 65% Fe supply, compressing Platts -$10-15/dmt and triggering Bear case.

Long-term value realisation (24-36 months) depends on decarbonisation commercial validation, where green iron premium emergence $5-10/dmt (conditional on China/EU emissions pricing >$50/tonne CO2 versus current $35-40) would justify $6bn Real Zero investment and create differentiation versus peers. Critical uncertainty involves whether first-mover advantage (3-5 year technology lead, 190MW solar operational Q1 FY26) translates to sustained premium or becomes stranded investment if peers leapfrog with proven technology within 3-5 years. Downside scenario crystallisation signals include: (1) China property sales decline accelerating beyond -30% YoY for 6+ consecutive months, (2) Iron Bridge production plateauing <4Mt/quarter by Q3-Q4 FY26, (3) Platts sustained <$100/dmt through June 2025 despite global ex-China demand stability, (4) Native title litigation loss (Yindjibarndi Federal Court February 2025) establishing $100-300m compensation precedent. Major uncertainties centre on China property sector structural versus cyclical decline (40% Bear/Severe combined probability reflects elevated tail risk), Iron Bridge technical execution (air classification yield, C1 cost trajectory), and decarbonisation ROI (market adoption <50% probability given Energy division commercial failures precedent).

Fortescue operates a vertically integrated iron ore mining business model spanning extraction (Pilbara hematite 172Mt, Iron Bridge magnetite 7Mt), rail transport (760km network, 210Mt capacity), port operations (Port Hedland 40% berth allocation), and shipping (10 VLOCs, 1.25bn revenue). Revenue generation follows commodity price-taker dynamics with 96% derived from iron ore sales to Chinese steel mills (>90% customer concentration), realising 83-87% of Platts 62% index for hematite (58% Fe grade discount) and 114% premium for magnetite (67% Fe quality advantage). Cost structure comprises 69% variable components (mining consumables, shipping, royalties 7.5% revenue) and 25% fixed infrastructure ($1.7bn baseline: rail maintenance, port facilities, corporate overhead), generating extreme operating leverage (DOL 2.4x: $10/dmt Platts swing = $1.5bn EBITDA impact = 20% earnings volatility). Business model sustainability centres on maintaining #2 global cost position ($18.0/wmt C1 cost) through automation (60% autonomous fleet, Fortescue Hive AI operations centre) and Pilbara geology advantages (low 1.6x strip ratio, shallow deposits <150m), though fading predictably as resource depletion forces higher strip ratios 1.6x→1.9x by 2035. Strategic pivot toward decarbonisation (Real Zero 2030: $6bn commitment, 190MW solar, 95ML diesel reduction) aims to create green iron premium $5-10/dmt differentiation, though Energy division commercialisation failures (-$2.0bn losses, zero third-party Fortescue Zero sales) signal execution risk.

Competitive advantages score 5.7/10 (moderate moat, 5-7 year duration) combining three primary sources. Cost leadership ($18.0/wmt versus peer average $20 = 10% structural advantage) derives from operational factors (automation 60% fleet, predictive maintenance, route optimisation) and natural endowments (Pilbara geology: 58% Fe hematite, shallow deposits, low overburden). Infrastructure barriers include Port Hedland duopoly (BHP 60%, FMG 40% berth allocation = $15bn replacement cost, 8-10 year approval timeline for competing Anketell port) and freight proximity ($10-15/dmt advantage versus Vale Brazil-China 10,000nm route versus Pilbara-China 3,500nm). Scale economies spread $2.5bn fixed infrastructure (rail, port, corporate) over 198Mt production, though #4 global ranking (12.4% seaborne share versus Rio Tinto 21%, Vale 19%, BHP 16%) limits bargaining power. Moat erosion timeline is predictable rather than uncertain: resource depletion forces new hub development (Nyidinghu, Mindy South) with strip ratios 1.9-2.1x (versus current 1.6x) increasing C1 cost $18→$20-21/wmt by 2032-35, whilst peer automation adoption (BHP/Rio 50%→70% by 2030) closes technology gap ~$1-2/wmt. Decarbonisation investments ($6bn Real Zero) represent moat maintenance (avoiding obsolescence if emissions regulations tighten) rather than widening, with 3-5 year first-mover window before peer replication (technology diffusion inevitable, no patent exclusivity on solar/wind/battery deployment).

Management assessment scores 6.3/10 overall, bifurcated by division: Metals operations 8.5/10 credibility (FY23-25 shipment guidance 95%+ achievement, C1 cost targets consistently met, automation deployment on schedule) versus Energy division 2/10 (Arizona/PEM50 cancelled post-FID after $500m+ spent, Fortescue Zero -$659m EBITDA loss with $144m revenue = 0.9% of consolidated). Founder Dr. Andrew Forrest's 21-year tenure (since 2003 founding) and 36% Tattarang ownership ($17bn stake) provides strategic continuity and aligned incentives, eliminating principal-agent risk. However, succession risk elevated (Forrest age 63, no disclosed plan; Metals CEO Dino Otranto appointed 2023 = 2 years tenure; Energy CEO Mark Hutchinson 2022 = 3 years) creates thin bench below founder. Capital allocation demonstrates value-creative discipline in Metals (Iron Bridge $3.5bn investment delivered despite 3-year delay, dividend policy 65-70% payout maintained through cycle = $6.5bn returned FY20-25) but value-destructive behaviour in Energy ($3.5bn invested with -$2.0bn losses = -57% return). Board quality adequate (Lead Independent Director Mark Barnaba 15 years, diverse NEDs including operational expertise) though governance questions arise from Energy division strategic pivots (Arizona/PEM50 cancellation signals poor project screening or market reality check). Compensation reasonable ($163m SBC FY25 = 1.0% revenue, in-line with mining sector 0.8-1.2% range) but lacks explicit ROIC/FCF targets, primarily revenue/production-linked with potential for value-destructive growth chasing.

FY25 results (year ending June 30, 2025) reflect late-cycle commodity trough dynamics, with revenue declining 15% to $15.5bn despite volume growth +4% to 179Mt shipments. Platts 62% iron ore index compression from $119/dmt (FY24) to $102/dmt (FY25) drove 14% price headwind, mechanically flowing through to revenue given 90% commodity price pass-through in business model. Hematite realization deteriorated from 87% to 83% of Platts index (grade discount widening as product mix shifted toward lower 58% Fe ore), whilst Iron Bridge magnetite maintained 114% premium pricing (67% Fe quality advantage) on initial 7Mt production ramp. EBITDA compressed 20% to $8.4bn (54.1% margin versus 60% historical average) as operating deleverage (fixed costs $1.7bn persistent) and hematite mix shift offset cost discipline achievements. Notably, C1 cost improved $18.24→$17.99/wmt (first decline since FY20), validating $500m automation investment ROI through 60% autonomous fleet deployment and Fortescue Hive AI-driven predictive maintenance. Energy division losses widened to -$659m EBITDA (7.8% drag on consolidated Metals EBITDA $9.1bn) following Arizona/PEM50 project cancellations post-FID, with Fortescue Zero manufacturing generating only $144m revenue (0.9% of total) against elevated R&D spend.

Balance sheet strength remained exceptional despite earnings compression, with net debt declining to $1.1bn (0.1x EBITDA, down from $1.5bn FY24) as free cash flow generation $3.1bn (estimated, 46% EBITDA conversion) exceeded dividend payments $3.4bn and sustained elevated capex $4.4bn. Cash balance $4.3bn provides substantial liquidity buffer, with interest coverage 23x (EBITDA $8.4bn ÷ interest expense $370m) eliminating refinancing risk. Management commentary emphasised "strategic pivot" in Energy division toward policy-supported jurisdictions (Norway Holmaneset, Brazil Pecém projects) following Arizona/PEM50 cancellations, interpreted as acknowledgment of over-ambitious prior commitments. FY26 guidance (announced August 2024) projects flat shipments 195-205Mt versus FY25 198Mt, signalling management capitulation on near-term recovery narrative and conservative positioning on China demand trajectory. C1 cost guidance $17.50-18.50/wmt maintains discipline, whilst Iron Bridge ramp trajectory targets 11Mt annualized Q1 FY26 (on track based on 7Mt FY25 exit rate). Capital allocation priorities confirmed: sustaining capex $2.3bn, decarbonisation $1.0-1.2bn through FY30, dividends 65-70% payout ($1.10-1.30 DPS range), with zero buyback activity despite authorization (missed value creation opportunity at $15-19 share price FY22-24 below intrinsic value).

Revenue trajectory models 2.6% CAGR (FY25→Terminal) driven by two-phase dynamics: near-term recovery FY26-28 (+12.6%, +8.6%, +5.3% annual growth) followed by post-FY28 plateau (flat to -1.5% as infrastructure capacity ceiling 210Mt enforced). Iron Bridge magnetite ramp from 7Mt (FY25) to 19Mt (FY28 base case, 86% of 22Mt nameplate capacity) contributes $2.4bn incremental revenue at 114% Platts index premium, whilst hematite stabilises at 172Mt optimal blend level. Platts iron ore price mean reversion assumption ($102/dmt FY25 → $108 FY26 → $112 FY27 → $115 FY28) reflects 5-year historical average normalization, mechanically flowing through to revenue via 90% commodity price pass-through. Hematite realization improves modestly from 83% trough (FY25) toward 86% (FY28) as grade discount normalizes, though structural 58% Fe disadvantage versus 62% Platts reference persists. Shipping revenue tracks iron ore volumes at ~7.5% of total ($1.25bn FY25 → $1.44bn FY28), whilst Other revenue (Fortescue Zero manufacturing, ancillary services) grows conservatively $230m → $320m reflecting limited commercial traction to date.

Margin progression models trough-to-recovery inflection, with EBITDA expanding from 54.1% (FY25) to 59.8% peak (FY28) before normalizing to 58.2% terminal. Operating leverage (DOL 2.4x) amplifies revenue growth $15.5bn→$20.0bn (+$4.5bn) into EBITDA expansion $8.4bn→$12.0bn (+$3.6bn) as fixed costs $1.7bn spread over larger base. Gross margin recovery from 60.5% (FY25) to 64.1% (FY28) driven 70% by Platts price recovery and 30% by Iron Bridge scale benefits (unit costs declining $36→$30/wmt as capacity utilization rises 32%→86%). Post-FY28 compression to 59% EBITDA margin reflects competitive equilibrium assumptions: peer automation adoption closing cost gap, resource depletion increasing strip ratios, and decarbonisation D&A drag ($2.38bn→$2.54bn on $6bn asset base) offsetting diesel savings. Terminal margin 58.2% sits conservatively below historical 60% average and FY21-22 super-cycle peak 65%, appropriately modeling CAP fade and margin normalization. Tax rate declines from 32% (FY26) to 30% (terminal) as foreign operations wind down and Australian Metals dominates, with effective rate benefiting from capital allowances and R&D credits.

Free cash flow generation improves from $3.2bn (FY26) to $5.7bn (terminal) as capex normalizes from peak $4.4bn (FY27, decarbonisation $1.2bn + Iron Bridge completion $0.5bn + sustaining $2.3bn) toward steady-state $2.9bn (sustaining $2.3bn + decarbonization maintenance $300m + exploration $300m). FCF/EBITDA conversion rises from 33% (FY26, elevated capex phase) to 51% (terminal, mature operations) with working capital providing modest tailwind as Iron Bridge inventory normalizes (12.0%→11.0% of revenue). Key assumptions underpinning forecasts: WACC 9.8% (cost of equity 12.2% via beta 1.19, cost of debt 5.5% post-tax), terminal growth 0% (volume ceiling 210Mt, China steel demand plateau), terminal capex $2.9bn (15% of revenue, industry norm for sustaining + decarbonization maintenance). Sensitivity analysis indicates Platts ±$10/dmt drives ±$2.00-2.50/share value impact (53% of variance), whilst WACC ±100bps creates ±15% valuation swing ($16.10-$23.50 range). Terminal value represents 48% of enterprise value (LOW DEPENDENCY versus 60% threshold), providing confidence in DCF reliability given strong near-term cash flow visibility.

DCF & Relative Valuation

Base case DCF produces $19.45/share fair value using 10-year explicit forecast (FY26-FY35) discounted at WACC 9.8%, with terminal value calculated via hybrid methodology (50% perpetuity growth at 0% + 50% EV/EBITDA multiple 8.0x sector median). Terminal value $73.4bn (undiscounted) represents 48% of enterprise value after discounting—LOW DEPENDENCY versus 60% threshold, validating DCF reliability given strong near-term cash flow visibility ($17.3bn PV Years 1-5 = 52% of total). Probability-weighted DCF $16.20/share incorporates scenario risk adjustment: Base case $19.45 (60% probability), Bear case $12.50 (30% probability, Platts $92/dmt sustained low prices), Severe case $7.81 (10% probability, China demand -5% structural decline). This 17% haircut versus Base reflects 40% combined downside probability (Bear + Severe), capturing China concentration risk (>90% sales) and Iron Bridge execution uncertainty (3-year delay history, 19Mt base versus 22Mt nameplate). Relative valuation via trading multiples: current 8.0x EV/EBITDA (FY25 $8.4bn) aligns with peer median 8.2x (BHP 8.6x, Rio Tinto 8.8x, Vale 7.3x), justified by #2 cost position offset by 58% Fe grade discount. P/E 15.7x (current) trades at premium versus peer median 10.8x despite #4 scale, reflecting ROIC 30% leadership (peer median 28%) though compressing to 20% terminal as CAP fades.

Scenario Analysis

Three-scenario framework captures commodity price cyclicality and execution risk with explicit probability weighting. Base case (60% probability, $19.45): Platts normalizes to $115/dmt mid-cycle, Iron Bridge achieves 19Mt (86% of nameplate), China steel production plateaus at 1.01bn tonnes, EBITDA margin recovers to 59.8% by FY28. Bear case (30% probability, $12.50): Platts sustained at $92/dmt (oversupply from Simandou or demand weakness), Iron Bridge ceiling 16Mt (technical constraints), China flat production, EBITDA margin 48%. Severe case (10% probability, $7.81): Platts <$85/dmt (China property structural collapse -5% steel demand), Iron Bridge uneconomic <15Mt, volume cuts to 180Mt, EBITDA margin 40%. Expected value $16.20 (probability-weighted) sits 17% below Base case, reflecting asymmetric downside risk (1.4:1 upside/downside ratio). 90% confidence interval $15.66-$21.18 (±15% band) derived from reliability score 86.6/100 (HIGH tier), incorporating low terminal dependency 48% and excellent data quality 9/10 for DCF inputs.

Market Pricing Dynamics

Current price $22.32 AUD ($14.49 USD equivalent at 1.54 exchange rate) trades 21% above dynamic weighted fair value $18.42 USD, creating material valuation gap requiring reverse DCF analysis to diagnose. Market pricing implies 35-38% ROE versus model's 28% base case, and/or revenue CAGR 5-6% versus model's 2.6%, neither sustainable given competitive dynamics (CAP fading 5-7 years, volume ceiling 210Mt enforced, terminal 0% growth physically constrained). Reverse-engineering the premium: at $22.32 AUD, market discounts either (1) Platts sustained $125-130/dmt (super-cycle pricing 15-20% above mid-cycle $115) for 5+ years, or (2) Iron Bridge achieving 22Mt nameplate with sustained 120% index premium (versus base 19Mt at 114%), or (3) terminal growth 1-2% (defying infrastructure capacity limits). Reality assessment suggests these assumptions optimistic given China property crisis deepening (construction -30% YoY = structural not cyclical), Iron Bridge 3-year delay history signaling execution risk, and physical volume ceiling 210Mt requiring $5-10bn Anketell port investment (2035+ timeline) for expansion.

Three behavioral and structural drivers sustain the mispricing. First, anchoring bias to FY21-22 super-cycle performance (Platts $140-160/dmt, EBITDA margins 65%, ROIC 36%) creates recency effect where investors extrapolate peak conditions despite mean reversion evidence (current Platts $102 = 13% below 5-year average $115, not temporary dip). Second, yield-seeking mandate flows from Australian retail investors (4.9% dividend yield versus ASX 200 average 4.2%) and income-focused superannuation funds create structural bid, with dividend sustainability perception lagging underlying commodity cycle deterioration (payout ratio 77.5% FY25 versus 65-70% target = unsustainable at trough earnings). Third, passive ETF rebalancing (MSCI Australia Materials 18% FMG weight) and index-tracking mandates provide mechanical buying regardless of valuation, with quarterly rebalancing flows (estimated $200-300m) dampening price discovery. These forces exhibit mixed stability: anchoring bias temporary (12-18 month catalyst = earnings normalization when FY26-27 results confirm margin compression if Platts fails to recover), yield flows durable but vulnerable to dividend cuts (probability 20-30% if Bear case materializes), ETF flows structural but reversible on index methodology changes.

Convergence catalysts with probability-weighted timing: Primary catalyst (probability 65%, horizon 12-18 months) involves earnings normalization when FY26-27 results demonstrate margin compression to 56-58% range if Platts recovery stalls <$110/dmt, triggering analyst downgrades and yield recalculation (dividend sustainability questions if payout ratio exceeds 80%). Early warning signal: Q2 FY26 production update (January 31, 2025) showing Iron Bridge <4Mt quarterly run-rate or Platts 3-month average <$105 would accelerate convergence timeline to 6-9 months. Secondary catalyst (probability 40%, horizon 18-24 months) involves interest rate regime change if RBA cuts 50-100bps (currently 4.35% cash rate), reducing relative yield appeal and triggering rotation from high-payout cyclicals toward growth equities. Tertiary catalyst (probability 25%, horizon 24-36 months) involves China property crisis escalation if construction starts decline accelerates beyond -40% YoY, forcing consensus recognition of structural (not cyclical) steel demand plateau and triggering sector-wide multiple compression. Monitor China property sales data monthly, Australian 10-year bond yields (convergence if spread <200bps versus dividend yield), and peer group relative performance (BHP/Rio outperformance signals quality flight).

China demand concentration (>90% sales exposure) represents primary tail risk, with property sector crisis (construction -30% YoY, sales -24% YoY October 2024) signaling potential structural decline rather than cyclical weakness. If property sector contracts -20-30% structurally (representing 30-40% of steel demand = 300-350Mt/yr), China steel production could decline from 1.01bn tonnes plateau to 0.95-0.96bn tonnes, triggering Bear case (30% probability, $12.50 fair value) or Severe case (10% probability, $7.81). Combined 40% downside scenario probability may underweight structural risk given property sector represents 30-40% household wealth (Evergrande, Country Garden bankruptcies = $5-10trn wealth destruction). Mitigation strategies limited: geographic diversification <5% non-China sales (India/Japan freight disadvantage versus NMDC domestic, Vale Atlantic basin), product differentiation via Iron Bridge magnetite insufficient (10% of volumes), cost position $18/wmt enables survival at Platts $90 but not growth. Early warning signals: China property sales decline accelerating beyond -30% YoY for 6+ consecutive months, steel production <950Mt/yr, seaborne import volumes declining >5% YoY.

Iron Bridge execution risk stems from 3-year delay history (FY25→FY28 nameplate) and novel air classification technology unproven at 22Mt scale. Base case assumes 19Mt achievement (86% of capacity) by FY28, but Bear case models 15-16Mt ceiling (30% probability) if processing yield remains <85% or C1 cost exceeds $32/wmt (versus $30 target). Value impact: -$300-400m EBITDA annually = -$1.50-2.00/share DCF reduction. Mitigation: blend magnetite with hematite to optimize grade/margin trade-off, cost optimization via scale benefits, reduce market expectations to 16Mt versus aspirational 22Mt. Monitor Q2-Q4 FY26 quarterly production (target >4.5Mt/qtr = 18Mt annualized), air classification yield % (target >88%), and C1 cost trajectory (target <$30/wmt by FY28). Decarbonisation stranded investment risk (15% probability, -$1.00-1.30/share) arises if $6bn Real Zero commitment fails to generate commercial returns—green iron premium absent (<50% probability given market adoption uncertainty), peer leapfrog with proven technology within 3-5 years (technology diffusion inevitable), or equipment reliability <90% uptime increases costs versus diesel baseline. Energy division precedent (-$2.0bn losses, zero Fortescue Zero third-party sales) validates execution skepticism on adjacent market entry outside core mining competency.