Vale S.A. (NYSE: VALE) — The Low-Cost King of a Commodity Whose Supply Cycle Is Finally Turning

Report date: 2026-06-19 · Price: $15.42 (Jun-18-2026 close) · Market cap: ~$65.8B · Shares: 4.269B common (+12 golden) · EV: ~$80–84B · FY-end: Dec 31 · Filer: Brazilian foreign private issuer (20-F/6-K, IFRS, USD) · CIK: 0000917851

⚡ Claude’s Take

This block is the author’s own independent opinion and general information only — not investment advice. The analysis in sections 1–15 below is deliberately position-free and carries no price target.

Verdict: HOLD / not-a-short / accumulate-on-weakness toward ~$11–13. Medium conviction. Vale is a genuinely low-cost, scaled iron-ore oligopolist — a real Greenwald cost-and-scale advantage that shows up in a sub-$24/t C1 cash cost and bottom-of-the-curve economics. But it is also a price-taker selling an undifferentiated commodity, with no demand-side moat, and the stock has already done the easy work: it is up ~80% in twelve months, sits just ~13% off its one-year high, and is being marked at ~5.5x EV/EBITDA on near-peak-of-recovery iron-ore prices (~$100–107/t). You are no longer buying it cheap; you are buying a recovered cyclical at a full-ish point in its own cycle, right as Simandou — the largest new iron-ore supply in decades — begins landing high-grade tonnes into a structurally plateauing Chinese steel market. That is the textbook Marathon late-cycle setup: a decade of supply discipline and high returns now meeting a wall of new capacity.

What keeps this a HOLD rather than an AVOID is that the price is fair, not euphoric, the ~6.6% dividend-plus-buyback yield pays you to wait, the two great overhangs (the Mariana settlement and the politicized CEO succession) have actually cleared, and copper is a credible second leg. What keeps it from being a BUY here is that the marginal news from here is more likely supply-driven downside (Simandou ramp, China peak-steel) than another leg of re-rating, and the Brazil/state-interference/dam-litigation discount is real and recurring. Framing (evidence-based, from the factor read): this is a Brazil + mining + anti-dollar EM cyclical (loadings: Industry-Mining +1.24, Country-Brazil +0.59, USDollar −0.53, Quality −0.20), low-quality, mildly value, not crowded momentum and not currently a falling knife — it already fell and recovered. The right way to own iron ore is at the bottom of the cycle being paid the dividend to wait, not after a +80% year. Bullish flip: sustained iron ore >$110 on Chinese demand re-acceleration or Simandou delays, plus delivery on copper growth → the low-cost producer compounds the yield. Bearish flip: iron ore breaks below ~$85 as Simandou ramps into soft China demand, and/or a politically-driven dilutive capital event. Tag: the best house on a street where they’re about to break ground on a thousand new ones.

📈 Stock Price Action — Five-Year Event Map

Factual price history (NYSE ADR, unadjusted close, 5-year daily price history). Price moves are FACT; attributed drivers are INTERPRETATION. No recommendation, no price target.

Vale’s ADR has round-tripped a full commodity cycle over five years: from a ~$22.94 post-COVID iron-ore-boom peak (Jul-2021) down to an ~$8.26 trough (Apr-2025) — a ~64% peak-to-trough draw-down — and back up ~+87% to $15.42 today, ~13% below its April-2026 one-year high of $17.82. The stock trades below its 21- and 50-day EMAs ($15.88 / $16.07) but comfortably above its rising 200-day EMA ($14.30): a name consolidating after a powerful year, not breaking down. 52-week range roughly $8.6–$17.8; beta ~0.87.

Cycle narrative. (1) The 2020–21 leg was a pure commodity super-spike — Chinese stimulus and supply tightness drove iron ore to a record, and Vale, the lowest-cost major, levered it into record cash and a ~$13.5bn dividend year. (2) The Evergrande-led China property crisis cut that in half almost overnight — the single clearest illustration that Vale’s fortunes are a derivative of Chinese construction. (3) A 2022 relief rally on Ukraine-driven supply fear and reopening optimism faded into (4) a two-year range as China’s recovery underwhelmed and the market digested the Mariana/Brumadinho overhangs. (5) Late-2024/early-2025 marked the cycle low: soft iron ore, a genuinely ugly state-influenced CEO succession, and a strong dollar combined to push the ADR under $9. (6) The past twelve months reversed all of it — iron ore stabilized above $100, Vale posted record 336Mt output, the Mariana settlement and CEO question were resolved, and a weak dollar plus copper strength added EM-commodity fuel for a ~+116% trough-to-peak run. (7) The recent ~13% pullback reflects iron ore drifting back toward $100 as the Simandou supply wave and China “peak steel” re-enter the narrative.

1. Executive Summary

Vale is the world’s largest iron-ore producer and one of the lowest-cost, alongside Rio Tinto and BHP — a genuine cost-and-scale advantage in a structurally cyclical, no-demand-moat commodity. Iron Ore Solutions is ~79% of revenue (iron ore 65%, pellets 11%); Vale Base Metals (VBM) — copper and nickel, branded “Energy Transition Materials” — is ~21% and the diversification/growth leg. FY2025 revenue was $38.4bn with EBITDA of $14.4bn (37.6% margin); attributable net income was a charge-depressed $2.35bn (vs $6.2bn in 2024 and $22.4bn at the 2021 peak), hit by a ~$4.6bn impairment (mostly Canadian nickel) plus FX and a 57% effective tax rate. Underlying run-rate earnings power, judged off Q1’26’s $1.89bn attributable net income, is closer to $6–8bn.

The investment tension is straightforward and almost mathematical. Vale earns its keep entirely through cost position and volume in a price-taking business — there is no pricing power, no customer captivity, no brand; the moat is “we can mine a tonne for ~$21–24 when the marginal producer needs ~$80–100.” That moat is real and durable (Carajás ore body, scale, infrastructure), and it produces excellent through-cycle returns: ROIC ran 47% at the 2021 peak and ~17–20% in 2023–24 before this year’s charge-driven dip to 8.8%. But the business is captive to the iron-ore price, which is captive to Chinese steel demand — now widely believed to have plateaued — and the supply side is turning: Simandou in Guinea, the largest new iron-ore source in decades (~120 Mtpa of high-grade ore at full ramp), began shipping in late 2025. This is the Marathon capital cycle inflecting from a decade of discipline to fresh oversupply.

Two long-standing discounts have genuinely cleared, which is most of why the stock is up ~80%: the Mariana/Samarco dam settlement (definitive, ~R$170bn over ~20 years, signed Oct-2024) and the politicized CEO succession (resolved with internal CFO Gustavo Pimenta, Oct-2024). What remains is a recurring state-interference discount (12 golden shares, Previ board influence, demonstrated Lula-government pressure) and live dam-litigation tail risk (the UK group action where Vale is on the hook for ~50% of BHP’s exposure; an ongoing criminal trial). Capital returns are a genuine strength — a ~6.6% dividend-plus-IoC yield, ~16% share-count reduction since 2020, an active buyback, and management signaling extraordinary dividends ahead.

Valuation is fair, not cheap: ~5.5x EV/EBITDA and ~2.0x book on near-peak-of-recovery prices, broadly in line with Rio/BHP and at a slight discount that the Brazil/dam/governance risk arguably justifies. The embedded expectation is roughly “iron ore holds ~$95–105 and copper grows” — neither heroic nor conservative. The skew from here is modestly downside-tilted: you have re-rated off the trough, the supply cycle is turning, and the dividend pays you to wait for a better entry rather than to chase.

2. Business Overview

Vale S.A. is a Brazilian multinational miner headquartered in Rio de Janeiro, founded in 1942 as the state-owned Companhia Vale do Rio Doce, privatized in 1997, and renamed Vale in 2009. It is the world’s largest producer of iron ore and iron-ore pellets and a top-tier producer of nickel and copper. It reports in two segments:

Iron Solutions (~79% of FY2025 revenue). This is the franchise. It comprises (per the FY2025 20-F segment table): iron ore fines $25.0bn (65.1% of revenue), iron-ore pellets $4.4bn (11.4%), and other ferrous products and logistics $0.7bn (1.9%). Vale produced ~336 Mt of iron ore in 2025 (a record at the flagship S11D mine, ~86 Mt) and guides to 335–345 Mt for 2026. Production is split across three “systems”:

• Northern System (Carajás, Pará): the crown jewel — the S11D mine and complex, hosting some of the world’s highest-grade hematite (~65% Fe), low impurities, open-pit, truckless. This ore commands a premium and underpins Vale’s low-emissions/“green steel” positioning.
• Southeastern and Southern Systems (Minas Gerais): older, lower-grade (~58–62% Fe) operations, more reliant on tailings dams (the source of the Mariana and Brumadinho disasters), partly displaced after 2019.

Vale also owns the logistics moat around the ore: dedicated railways (EFC, EFVM), the Ponta da Madeira and Guaíba/Itaguaí maritime terminals, and a fleet/affreightment book of Valemax very-large ore carriers that narrows its freight disadvantage versus Australia on the long haul to China. Roughly 60%+ of iron-ore revenue is China-destined.

Vale Base Metals / Energy Transition Materials (~21%). Nickel was 11.2% of FY2025 revenue ($4.3bn) and copper ~10% (plus by-product gold, silver, cobalt, PGMs). Copper assets are Salobo and Sossego in Carajás (low-cost, growing); nickel is mostly higher-cost Canadian sulphide (Sudbury, Voisey’s Bay, Thompson) plus Onça Puma (Brazil) and a now-minority stake in PT Vale Indonesia. In 2023–24 Vale sold 10% of VBM to Saudi Arabia’s Manara Minerals for ~$2.5bn, implying a ~$26bn enterprise value for the unit; Vale retains ~90%. Management (VBM CEO Shaun Usmar, ex-Barrick/Xstrata) has a “fix-the-base” cost-and-reliability mandate and an ambition to make VBM “IPO-ready by 2027.”

How it makes money: Vale sells a commodity at a market-set price (the 62% Fe index plus quality premia/discounts) and keeps the spread between that price and its cost to mine, process, and ship. There is essentially no recurring or contracted revenue in the software sense — volumes are sold under index-linked contracts and spot. Profitability is therefore the product of (price − unit cost) × volume, and price is exogenous. This is the defining feature of the business and of this memo.

3. Industry Dynamics

Structure: a concentrated oligopoly selling a commodity. Seaborne iron ore is dominated by four producers — Vale, Rio Tinto, BHP, and Fortescue — which together control ~70%+ of seaborne supply. Approximate 2025 volumes: Vale ~336 Mt, Rio (Pilbara) ~270–290 Mt, BHP ~250+ Mt, Fortescue ~190–197 Mt. On paper this is an attractive structure: high concentration, enormous capital and infrastructure barriers to entry (a new world-class mine plus rail plus port is a multi-decade, multi-tens-of-billions undertaking), and rational incumbents. In Greenwald terms the barriers to entry are genuine — economies of scale plus access to unique, low-cost ore bodies.

But the product has no demand-side moat. Iron ore is fungible within grade bands; steelmakers buy on price and quality, not loyalty. There are no switching costs and no brand. The oligopoly therefore competes on cost, and the industry’s profit pool is set by the marginal tonne — high-cost Chinese domestic mines and junior seaborne producers at ~$80–100/t. When the price is high, the low-cost majors earn spectacular margins (Vale’s 2021 EBITDA margin was 62%); when demand softens, the price falls toward the marginal cost and margins compress (37.6% in 2025). This is why the industry is “structurally good business, structurally bad pricing” — excellent assets, zero control over the selling price.

Demand: China, and the “peak steel” question. Iron ore demand is ~70%+ a bet on Chinese steel, which is ~50%+ a bet on Chinese construction/property. China’s property sector has been in a multi-year contraction since the 2021 Evergrande crisis, and the consensus has shifted toward “peak steel”: Chinese pig-iron output is widely believed to have plateaued and may now decline structurally, with provincial environmental output caps reinforcing the ceiling. India and Southeast Asia are growing steel consumers but are not large enough, soon enough, to fully offset a Chinese decline. The mid-2026 price action — 62% Fe drifting from ~$107 toward ~$100 on high Chinese port inventories and peaked pig-iron — is the market pricing this in real time.

Supply: the cycle is turning (the key structural risk). For roughly 2016–2024 the majors were a model of capital discipline — post-2015-crash, they prioritized debt reduction, brownfield debottlenecking, and shareholder returns over greenfield growth, which sustained high returns. That discipline is now ending on the supply side. Simandou in Guinea — developed by Rio Tinto/Chinalco (Simfer) and a Baowu-led Chinese consortium (Winning) — formally began operations in November 2025, with first cargo reaching China in January 2026; it targets ~60 Mtpa intermediate ramping to ~120 Mtpa of ~65% Fe high-grade ore by ~2028. This is the largest new iron-ore supply addition in decades, and crucially it is high-grade, directly eroding both the price floor and the high-grade premium that Vale’s Carajás ore enjoys. Layered on top of Vale’s own +20 Mt S11D expansion and Rio’s replacement tonnes, the seaborne market is adding supply into a plateauing-demand market. In Marathon’s framework this is the classic late-cycle inflection: a decade of high returns has finally attracted the capital that will compete those returns away.

Regulation and royalties. Brazil levies the CFEM mining royalty at 3.5% of net iron-ore revenue; iron-ore exports are not currently export-taxed. The binding regulatory reality for Vale is dam safety — post-Brumadinho, Brazil’s ANM mandated de-characterization (removal) of all upstream tailings dams, a multi-year, multi-billion-dollar program, and tightened the licensing/oversight regime. A periodic political risk is a windfall or export tax floated by the Lula government amid criticism of dividends-over-domestic-reinvestment; it is not enacted, but it is a recurring tail.

Verdict: a structurally mixed industry. The barriers to entry and concentration are genuinely high (good), but the product is a price-taking commodity tied to a single, plateauing demand center (China), and the supply cycle is inflecting toward oversupply just as demand peaks (bad). Net: a high-quality-asset industry at a structurally challenged point in its cycle. Good business, wrong part of the cycle.

4. Competitive Position

The moat is cost and scale — and it is real. Vale is at or near the bottom of the global iron-ore cost curve. Its C1 cash cost (ex-third-party purchases) was ~$20.7/t in Q3’25 and $23.6/t in Q1’26 (the increase driven mostly by a stronger Brazilian real, not operational slippage), with 2026 C1 guidance of $20.0–21.5/t. Against a 62% Fe price near $100 and a marginal producer cost of ~$80–100, that is a structural margin the marginal tonne cannot touch. The advantage has three sources, all genuine in Greenwald’s taxonomy:

1. A unique, world-class ore body (Carajás/S11D): ~65% Fe hematite with low contaminants, open-pit, enormous reserve life. You cannot replicate Carajás; you can only find another one (which is what Simandou is).
2. Economies of scale and integrated logistics: dedicated railways, captive ports, and a Valemax fleet that spread fixed infrastructure over ~336 Mt and narrow the freight gap to China. A junior producer has none of this.
3. A high-grade/low-impurity quality premium that is increasingly valuable as steelmakers decarbonize (higher-grade feed reduces coke and emissions). This is Vale’s most defensible differentiation, though it is partly threatened by Simandou’s similarly high-grade ore.

But the moat protects the cost position, not the price. This is the crucial distinction. Vale’s advantage guarantees it will be among the last producers standing and the most profitable at any given price — it does not give it any control over the price itself. When iron ore fell from $233 to $90 in 2021, Vale’s cost advantage did nothing to stop its earnings (and stock) halving. The moat is a relative-survival moat, not a pricing moat. A useful test: would Vale’s economics deteriorate without the moat? Yes — it would be a marginal, loss-making producer at $90 ore instead of a 35%+ EBITDA-margin one. So the moat is real and financially load-bearing. But it caps downside relative to peers; it does not create absolute pricing power.

Versus the majors. On cost, Vale, Rio, and BHP are roughly comparable bottom-of-curve producers (Vale C1 ~$21–24/t; Rio Pilbara C1 ~$23.5/wmt), with Fortescue slightly higher and lower-grade. Vale’s all-in cost (~$55/t in Q1’26) is higher than the Australians’ because of the longer freight haul from Brazil to China and Brazilian royalties/logistics — a permanent geographic disadvantage the Valemax fleet only partly offsets. Vale’s offsetting edges are grade (Carajás premium) and product flexibility (it can blend and shift between systems). Against Simandou, Vale loses some of its grade differentiation but retains scale, established logistics, and a 30-year head start.

Base metals: a different competitive picture. In copper, Salobo/Sossego are genuinely low-cost Tier-1 assets and a credible growth engine (copper production hit 102kt in Q1’26, the highest since 2017). In nickel, Vale is competitively disadvantaged: its Canadian sulphide nickel is higher-cost and has been structurally undercut by a flood of low-cost Indonesian laterite/HPAL nickel backed by Chinese capital — which is precisely why Vale took a ~$3.5bn impairment on those assets in late 2025. So VBM is two businesses: a competitively strong copper story and a competitively weak nickel one that management is trying to ring-fence.

Verdict: a durable but one-sided advantage. Vale has a real, financially material, durable cost-and-scale moat in iron ore and copper — but it is a moat that protects relative profitability in a price-taking business, not one that confers pricing power. It is the best house on the street; the street’s rent is set by China.

5. Growth History and Forward Opportunities

History: a volume-recovery and a price-cycle story, not a secular grower. Vale’s revenue has been a function of price far more than volume. Revenue ran $39.5bn (2020) → $54.5bn (2021 boom) → $43.8bn (2022) → $41.8bn (2023) → $38.1bn (2024) → $38.4bn (2025) — i.e., it is lower in nominal terms than five years ago, despite higher iron-ore volumes, because the iron-ore price normalized off its super-spike. Volumes, by contrast, have been a recovery story: iron-ore output collapsed from ~385 Mt (2018) to ~300 Mt after Brumadinho halted ~90 Mt of capacity, and has since clawed back to ~336 Mt (2025). So the “growth” of the past five years has been getting back to where it was — re-permitting and ramping displaced tonnes — not net expansion.

Forward opportunities, ranked by credibility:

1. Iron-ore volume to ~340–360 Mt (modest, credible). The Serra Sul +20 (S11D +20 Mt) project is 86% complete and starts up in H2’26; Capanema and Vargem Grande are ramping. Medium-term Vale targets ~340–360 Mt. This is real but incremental, and it adds supply into a softening price — volume growth that may be partly self-defeating on price.
2. Copper growth (the genuine secular leg). VBM copper production (~340–370 kt guided for 2025) has a long runway — Usmar has pointed to ~700 kt by ~2035 via Salobo/Sossego expansions and the Carajás copper district, plus Hu’u (Indonesia) longer-term. Copper is the one part of Vale with a structural-demand tailwind (electrification, grid, data-center power). This is the most attractive growth Vale has, and the rationale for the VBM minority sale and possible 2027 IPO.
3. High-grade/agglomeration & “green steel” (optional, slow). Vale’s “Mega Hubs” (briquettes, concentration plants, partnerships in the Middle East to produce low-emission metallics) are a real but early, capital-intensive bet on a decarbonizing steel value chain. Credible optionality, not a near-term earnings driver.
4. Nickel (negative growth / shrinkage). The realistic path here is cost-out, asset rationalization, and possibly partial exit, not growth — the market is oversupplied and Vale’s assets are high-cost.

Quality of growth. Modest and capital-intensive. Iron-ore volume growth is low-return-on-incremental-capital at the margin (it arguably worsens the industry’s price problem); copper growth is the high-quality piece but requires heavy capex (Vale’s total capex is rising toward ~$6.5bn) and multi-year build times. There is no high-return, capital-light growth here — this is a heavy-industry capital allocator, not a compounder.

Verdict: low-quality, price-dependent growth with one genuine bright spot (copper). Top-line growth is hostage to the iron-ore price; the credible volume growth is incremental and partly self-cannibalizing on price; copper is the only structurally attractive expansion, and it is years from moving the consolidated needle.

6. Financial Quality

Margins and their cyclicality. Vale’s economics are excellent at the top of the cycle and merely good at the bottom — the signature of a low-cost commodity producer. Operating margin: 56.6% (2021) → 40.6% (2022) → 38.0% (2023) → 29.2% (2024) → 29.5% (2025). EBITDA margin held ~37.6% in 2025 even as iron ore normalized — a testament to the cost position. EBITDA in absolute terms fell from $33.9bn (2021) to $14.4bn (2025), almost entirely on price. The key financial-quality point is that the margin floor is high (mid-30s% EBITDA at ~$100 ore) because of the cost advantage — Vale does not lose money in a downturn; it just earns less.

Returns on capital — the capital-cycle signature. ROIC: 46.8% (2021) → 27.9% (2022) → 20.4% (2023) → 17.6% (2024) → 8.8% (2025). ROE: 36% (2021) → 30% (2022) → 13% (2023) → 10% (2024) → 3.8% (2025). The 2025 collapse in returns is charge-driven (see QoE below), not operational; normalized through-cycle ROIC is mid-teens, comfortably above an ~8–10% cost of capital at mid-cycle prices and far above it at the top. This is the empirical proof the moat is real — but it is also a textbook Marathon mean-reversion: spectacular returns at the price peak compressing toward the cost of capital as the price normalizes and (prospectively) as new supply lands.

Quality of earnings — 2025 GAAP is depressed, a green flag. FY2025 attributable net income of $2.35bn dramatically understates earnings power. It was hit by: (i) a ~$4.6bn impairment charge (primarily ~$3.5bn on Canadian nickel assets on a lowered nickel price deck, plus a deferred-tax write-off); (ii) Mariana/Samarco provision movements and FX on real-denominated provisions; and (iii) a 57% effective tax rate (vs ~11–27% in normal years), inflated by the non-deductible/deferred-tax items. The cleaner read: Q1’26 alone delivered $1.89bn of attributable net income (+36% YoY), $9.26bn revenue, and $3.83bn adjusted EBITDA — annualizing to a ~$15–16bn EBITDA run-rate and ~$6–8bn of normalized earnings. So the headline P/E of ~37x is a trough-earnings artifact; normalized P/E is ~8–11x. This is a QoE green flag: GAAP is below cash/operating reality, not above it. (By contrast, the 2021–22 GAAP figures were flattered by the boom and should not be annualized either — normalize both ends.)

Cash flow. FY2025 operating cash flow was $8.8bn; capex $6.0bn; free cash flow ~$2.8bn. Notably, the $3.56bn of dividends paid in 2025 exceeded free cash flow — the gap was debt-funded, which is why net debt drifted up. This is acceptable at a cyclical trough in earnings (management is smoothing the dividend through a weak year) but is a watch-item if iron ore weakens further: the ~6.6% yield is not bulletproof if prices fall and capex stays elevated. At mid-cycle prices, FCF coverage of the dividend is comfortable (2023 FCF was $7.2bn); at trough prices it is not.

Balance sheet. Solid but carrying disaster baggage. At Dec-2025: cash $7.57bn, gross debt ~$21.8bn, conventional net debt $13.76bn, minority interest $0.84bn, equity ex-minority $33.5bn (book value ~$7.85/share, so P/B ~2.0x). Crucially, Vale manages to an “expanded net debt” figure that adds the Brumadinho/Samarco and de-characterization provisions and leases on top of conventional net debt: $15.6bn at end-2025, $17.8bn in Q1’26 (seasonally elevated), against a target band of $10–20bn with a midpoint goal of ~$15bn. This is the right way to think about Vale’s leverage — the dam liabilities are real, senior, cash-consuming obligations. Even so, ~$17.8bn of expanded net debt against ~$15bn of run-rate EBITDA is ~1.2x — investment-grade and manageable.

Verdict: economics that are genuinely good and improve with scale, wrapped around a price the company doesn’t control and a liability tail it created. The cost advantage gives a high margin floor and mid-teens through-cycle ROIC; the QoE is clean (GAAP currently understates earnings); the balance sheet is sound on an expanded-net-debt basis. The honest caveat is that the dividend is not fully covered at trough prices.

7. Capital Allocation

The record is mixed, and improving. Vale’s capital-allocation history contains both the best and the worst of mining capital allocation.

The worst: the dam disasters. Mariana (2015) and Brumadinho (2019) were, at root, capital-allocation and risk-management failures — under-investment in tailings-dam safety in pursuit of low-cost production, with catastrophic human and financial consequences (19 and 270 deaths respectively, and tens of billions in settlements and remediation). No capital-allocation assessment of Vale can omit that the company destroyed enormous value — and far more than value — through inadequate safety spending. The de-characterization program (removing all upstream dams) is the belated correction.

The best: disciplined returns and a shrinking share count. Post-2019, Vale has been a strong returner of capital. It has reduced its share count from ~5.1bn (2020) to ~4.27bn (2025), roughly −16%, via consistent buybacks — and crucially, much of that was repurchased in the $11–15 range, i.e., at sensible (cyclically depressed) prices, not at the top. It pays a variable dividend plus interest-on-capital (a Brazilian tax-efficient distribution), targeting a ~30% of (EBITDA − sustaining capex) minimum policy plus extraordinary distributions when leverage sits below the midpoint. The trailing yield is ~6.6%. In Q1’26 it distributed $2.7bn and repurchased ~5M shares, and management stated it is “increasingly confident” about extraordinary dividends and further buybacks at current commodity prices. This is genuinely shareholder-friendly, returns-focused capital allocation.

M&A and portfolio. The signature recent move — selling 10% of VBM to Saudi Arabia’s Manara Minerals for ~$2.5bn at a ~$26bn implied EV (2023–24) — was a smart crystallization of value in the base-metals unit at a high mark, bringing in a strategic partner without ceding control and funding copper growth. (The companion Engine No. 1 3% tranche falling through is a mild negative on outside-investor appetite, and the ~$26bn mark looks rich after the subsequent nickel impairment.) The progressive sell-down of PT Vale Indonesia (to ~34%) was regulatory-driven (Indonesian local-ownership rules) and sensible given the nickel headwinds. Net, the portfolio moves have been rational and value-aware.

Capex discipline. Capex is rising (~$6.0bn in 2025, guiding ~$6.5bn) as copper growth and the S11D expansion are funded, but Vale has not embarked on a value-destroying greenfield iron-ore spree — its iron-ore growth is brownfield and incremental. That restraint is itself a positive capital-allocation signal in a late-cycle industry.

Incentives and governance overlay. The compensation plan now includes a ROIC hurdle (~25% weight from 2024) alongside TSR and ESG metrics, and a CEO equity-holding requirement (36x fixed fee) — a meaningful improvement that aligns management with returns, not just volume. Insider ownership is negligible (a foreign-private-issuer norm; Form 4 activity is sparse and not a useful signal). The real governance issue is not the comp plan but who controls the board — see §8.

Verdict: improving, returns-focused capital allocation with a catastrophic legacy. Setting aside the dam disasters (a permanent stain and a continuing liability), the current capital-allocation regime — disciplined buybacks at sensible prices, a high variable dividend, a value-aware VBM partnership, brownfield-only iron-ore growth, and a new ROIC incentive — is good. The watch-items are dividend coverage at trough prices and the risk that state influence redirects capital toward political rather than economic ends.

8. Changes and Headwinds — Last Two Years

The two great overhangs cleared (the bull’s foundation).

• Mariana/Samarco settlement (Oct-25-2024): a definitive ~R$170bn settlement over ~20 years removed the single largest open-ended legal uncertainty. The liability is now quantified and amortizing — R$38bn already invested, ~R$100bn in installments to 2043, R$32bn in Samarco performance obligations — and captured in expanded net debt. Vale’s recorded Samarco provision was ~$2.6bn at end-2025.
• CEO succession resolved (Oct-1-2024): after a genuinely ugly, politicized process (see below), the board appointed internal CFO Gustavo Pimenta as CEO. Continuity and an operations/finance focus, rather than a political appointee, was the market-relief outcome.

Operational and financial momentum. Record 2025 iron-ore output (336 Mt); record Q1’26 copper (102kt) and strong nickel volumes (base-metals EBITDA +116% YoY in Q1’26); a ~$1.5bn upward revision to the 2026 FCF outlook (May-2026) on the iron-ore rally; and successive analyst upgrades (JPMorgan to Overweight, PT raised to $21 on Jun-12-2026; BofA naming VALE a 2026 top pick).

The headwinds.

• Simandou supply wave (the structural bear case): first ore shipped late-2025, ramping to ~120 Mtpa high-grade — pressuring both price and Vale’s grade premium for years.
• China “peak steel”: plateauing/declining Chinese pig-iron, high port inventories, environmental output caps; iron ore drifting toward $100 in mid-2026.
• Live dam litigation: the English High Court found BHP liable (Nov-2025) in the Mariana group action (>620k claimants, ~£36bn claimed), and Vale is contractually on the hook for ~50% of BHP’s liability — a real, unquantified incremental tail on top of the Brazilian settlement. The Brumadinho criminal trial of Vale and 16 executives began Feb-2026 and runs into 2027.
• State interference (recurring): the 2024 CEO fight exposed it; the Cosan strategic stake exit (Jan-2025) removed an anchor holder.
• Nickel impairment (late-2025): ~$3.5bn write-down confirming the structural weakness of the Canadian nickel franchise.

Verdict: the changes are net thesis-strengthening on the legacy overhangs but thesis-weakening on the forward cycle. The past two years removed the disaster-and-succession discount (which the stock has now banked, +80%), but the next two years introduce the supply-cycle and China-demand headwinds that the stock has not yet fully discounted. The good news is largely in the price; the structural challenge is ahead.

9. Risk Analysis (Risk Matrix)

Reading the matrix. The dominant risks are price (China demand) and supply (Simandou) — both High likelihood / High impact, and both pointing the same direction (down for iron ore). These are not idiosyncratic, diversifiable risks; they are the core of owning Vale. The dam-litigation and state-interference risks are the Brazil-specific overlay that justifies a structural discount to the Australian majors. The reassuring side: the cost position makes a catastrophic financial outcome very unlikely — Vale stays profitable and investment-grade even in a downturn; the tail risk that would impair the equity permanently is another operational disaster, which de-characterization is steadily (if slowly) reducing.

10. Valuation Discussion (Embedded Expectations)

Method. For a price-taking cyclical, EV/EBITDA and dividend/FCF yield are the right lenses; P/E is distorted (2025 GAAP is charge-depressed; 2021–22 was boom-flattered), and own-history percentile-rank screens are unusable here because common data-feed per-share inputs for VALE are garbled (e.g. sales/share of $1.84 vs an actual ~$9 and book/share of $1.64 vs ~$7.85, producing nonsensical multiples). We anchor on EV/EBITDA at a live-price-rebuilt EV, the dividend yield, and a scenario analysis keyed to the iron-ore price.

Live EV rebuild. At $15.42 and 4.269bn shares, market cap is ~$65.8bn. Adding conventional net debt ($13.76bn) and minority interest ($0.84bn) gives EV ≈ $80.4bn; using the more conservative expanded net debt ($17.8bn, including dam provisions) gives EV ≈ $84.4bn. (A common data-feed EV of ~$70.7bn uses a stale ~$55.6bn FY-end market-cap snapshot and understates EV by ~$10bn; use the live rebuild.)

Current multiples:

• EV/EBITDA: $80–84bn / $14.45bn FY2025 EBITDA = ~5.6–5.8x trailing; on a ~$15–16bn run-rate (Q1’26 annualized) = ~5.1–5.5x forward.
• P/B: ~2.0x (book ~$7.85/share).
• Dividend yield: ~6.6% trailing; FCF yield ~4% (2025, trough) to ~10%+ (mid-cycle).
• Normalized P/E: ~8–11x on ~$6–8bn normalized earnings.

Peer comparison. Rio Tinto and BHP trade at ~5.0–5.8x EV/EBITDA; Fortescue ~4.5–5x (lower-grade, single-commodity); Vale at ~5.5x sits roughly in line to a slight discount. That modest discount is reasonable — it reflects Brazil/state-interference risk, the dam-litigation tail, the freight disadvantage, and the nickel drag, partly offset by the copper optionality and the high grade. Vale is neither anomalously cheap nor expensive versus its peer group; it is priced as what it is — a slightly-higher-risk member of the low-cost-major club.

Embedded expectations. At ~5.5x forward EV/EBITDA and ~$15–16bn EBITDA, the market is underwriting roughly: iron ore holds ~$95–105/t through-cycle, Vale’s volumes grow modestly to ~340–360 Mt, copper delivers incremental growth, and the dam liabilities amortize without major new surprises. That is neither a bull nor a bear assumption — it is a “muddle-through mid-cycle” expectation. What the market is arguably under-weighting is the asymmetry of the supply cycle: Simandou is a multi-year, ~120 Mtpa high-grade overhang that did not exist in prior cycles, and “peak China steel” caps the demand offset. What it may be correctly pricing is that Vale’s cost position protects it relative to peers and that the cleared overhangs justify the re-rating off the trough.

What must be true for the price to work (and to fail). For ~$15.42 to be cheap, you need iron ore to hold ≥$100 despite Simandou ramping into peak-steel China — i.e., you need Chinese demand to surprise up, or Simandou to disappoint/delay, or India/SE Asia to offset faster than expected. For it to be expensive, you need iron ore to mean-revert toward the ~$80–90 marginal cost as new supply lands — at which point EBITDA falls toward ~$11–12bn, the multiple de-rates on a worsening cycle, and the stock retraces toward its 2024–25 lows. Both are plausible; the supply-cycle evidence tilts the probability modestly toward the second.

Scenario analysis (3-year, illustrative — not price targets):

• Bear (~30%): Iron ore reverts to ~$80–90 as Simandou ramps into soft China demand; EBITDA ~$11–12bn; dividend trimmed; multiple de-rates to ~4.5–5x on a worsening cycle. Equity well below current levels, cushioned by the ~6%+ yield and the cost-position margin floor. Dead-money-to-down, not impairment.
• Base (~45%): Iron ore holds ~$95–105; volumes to ~345 Mt; copper grows; EBITDA ~$14–16bn; multiple holds ~5.5x; the ~6.6% yield + modest buyback carries the return. Roughly flat-to-modestly-higher total return, dividend-driven.
• Bull (~25%): China demand re-accelerates and/or Simandou delays; iron ore >$110; copper delivers and VBM IPOs at a premium mark; EBITDA ~$17–19bn; multiple re-rates toward ~6x. Materially higher, with the low-cost producer levering the upside.

The skew is modestly downside-tilted from spot — the bear and base together (~75%) cluster at-or-below current levels, the bull requires the supply/demand cycle to cooperate against the prevailing evidence. The ~6.6% yield is the ballast that makes the downside “pay-you-to-wait” rather than “value-trap.”

11. Variant Perception

Consensus. Wall Street is constructively positioned: a Buy/Moderate-Buy tilt, an average price target lagging in the ~$16.6–17.3 range with the marginal revisions rising (JPMorgan to $21, BofA top-pick). The consensus story is “low-cost major, cleared overhangs, ~6.6% yield, copper optionality, iron ore stabilizing above $100” — broadly the bull-to-base case. The factor tape agrees it is not a contrarian/abandoned name: rs_12m +79.8, ~13% off its high, above its 200-EMA — a recovered, in-favor cyclical.

The strongest bull case. Vale is the lowest-cost producer of the world’s most important industrial commodity, at ~5.5x EBITDA and a ~6.6% yield, after its two biggest discounts (dam settlement, CEO) have cleared, with a genuine copper-growth and VBM-monetization kicker and a disciplined buyback shrinking the share count at sensible prices. If iron ore merely holds ~$100, you collect a high-single-digit total yield with re-rating optionality; if China stimulates, you have ~2x operating leverage to the upside. The cost moat caps the downside relative to peers. This is “own the best low-cost cyclical at a fair multiple and get paid to wait.”

The strongest bear case. You are buying a price-taking commodity producer after a +80% year, near its highs, at a multiple that embeds mid-cycle iron ore — exactly as the largest new supply in decades (Simandou) begins landing high-grade tonnes into a Chinese steel market that has structurally peaked. This is the Marathon late-cycle trap: high prior returns have finally attracted the capital that competes them away, and the marginal tonne is about to reset the price toward ~$80–90, where Vale’s EBITDA falls ~25% and the multiple de-rates on a deteriorating cycle. Layer on a real Brazil/state-interference discount, a live UK litigation tail, an undividend-covered payout at trough prices, and a structurally broken nickel franchise. The cleared overhangs are why the stock is no longer cheap; the supply cycle is why it should not be.

The 3–5 assumptions that matter most:

1. The iron-ore price path (function of Chinese steel demand and seaborne supply) — dominates everything.
2. Simandou’s ramp pace and grade impact — the swing factor on the supply side and on Vale’s premium.
3. Chinese steel demand trajectory (“peak steel” vs. a stimulus-driven reacceleration; India/SE Asia offset).
4. Copper delivery and VBM monetization — whether the diversification leg materially de-commoditizes the story.
5. Brazilian political/state risk — whether governance stays economic or turns political (windfall tax, capital direction, board interference).

What would falsify each side. Bull falsified if iron ore breaks and holds below ~$85 as Simandou ramps, with no Chinese demand offset and a trimmed dividend. Bear falsified if Chinese demand re-accelerates (or Simandou materially disappoints), iron ore holds ≥$105, and copper growth plus a VBM IPO re-rate the multiple. The factor-positioning read sharpens the variant perception: the tape says consensus has already moved to the bull/base view (recovered, in-favor, near highs) — which means the easy money on “overhangs clearing” is made, and the marginal buyer is now implicitly betting on the commodity, not the re-rating. The under-appreciated risk is therefore supply-side and on the downside; the under-appreciated reward requires a China surprise. Consensus is offsides not on the business quality (it has that right) but on the timing within the cycle.

12. Fact vs. Interpretation

13. Open Questions

1. What is the real through-cycle iron-ore price once Simandou is fully ramped and Chinese steel has settled into its post-peak trajectory — $80, $90, or $100? This single number drives the entire valuation.
2. How fast and how high-grade does Simandou actually ramp, and how much of Vale’s quality premium does it erode? (Guinea execution risk cuts both ways.)
3. Will the VBM IPO (“IPO-ready by 2027”) actually happen, at what mark, and does the ~$26bn 2024 valuation survive the nickel impairment? This is the cleanest catalyst for de-commoditizing the story.
4. Is the ~6.6% dividend safe through a trough, or will a sub-$90 iron-ore year force a cut (as the uncovered 2025 payout warns)?
5. How durable is the post-2024 governance settlement — does the board stay independent of the Lula government through the 2026 Brazilian election cycle, or does state pressure (windfall tax, board interference, capital direction) resurface?
6. What is the incremental cost of the UK group action on top of the Brazilian settlement, and when does it crystallize?
7. Can copper growth (toward ~700kt) actually move the consolidated needle within the investment horizon, or is it a 2030s story?

14. What Must Be True

Bull case — what must be true:

1. Iron ore holds ≥~$100/t through-cycle — i.e., Chinese demand stays resilient (or India/SE Asia offsets the decline) and Simandou ramps slowly enough that net seaborne supply stays balanced.
2. Vale executes its modest volume growth (to ~345–360 Mt) and copper growth (toward ~700kt) on budget, and monetizes VBM at a healthy mark.
3. The dividend-plus-buyback yield (~7–9% total) is sustained, and governance stays economic (no value-destroying state interference).

Falsification test: iron ore breaks and holds below ~$85 for two-plus quarters as Simandou ramps, the dividend is trimmed, and EBITDA falls toward ~$11bn — the bull case is broken.

Bear case — what must be true:

1. The seaborne market tips into oversupply as Simandou (+Vale’s own +20Mt, +Rio replacement) lands into a structurally peaked Chinese steel market; iron ore mean-reverts toward ~$80–90.
2. EBITDA compresses ~25% toward ~$11–12bn and the multiple de-rates on a deteriorating cycle; the uncovered dividend is cut.
3. The Brazil/state/litigation discount persists or widens (UK litigation cost, windfall-tax/political risk).

Falsification test: Chinese steel demand re-accelerates (or Simandou materially disappoints), iron ore holds ≥$105, copper delivers, and a VBM IPO prints a premium mark — the bear case is broken and the multiple re-rates.

Synthesis. The bull and bear cases are not really about Vale’s business quality — both sides concede it is the (or a) lowest-cost major with a real moat. They are about where you are in the commodity cycle and the supply curve. Vale is a fair-to-attractive way to own the low-cost end of iron ore, but it is being bought after the overhangs cleared and the stock re-rated +80%, as the largest new supply in decades begins to land. That is why the disciplined position is to own it at the bottom of the cycle being paid the dividend to wait — and to accumulate on the weakness the supply cycle is likely to bring, not to chase it near its highs.

15. Source Appendix

See Appendix B (Source Appendix) below for the full source list with URLs and access dates.

This is an independent analyst’s published research. The institutional analysis (sections 1–15) is deliberately position-free and contains no buy/sell recommendation or price target; the single clearly-labeled subjective view is the author’s own Claude's Take opening block.

APPENDIX A — Standard Diligence Questionnaire — Vale S.A. (NYSE: VALE)

Supplemental to the research memo. Fact / Interpretation / Assumption labeled where it matters. As of 2026-06-19.

General

What thoughtful questions have other investors asked about this company? The recurring institutional questions: (1) Where is the real through-cycle iron-ore price once China steel peaks and Simandou ramps? (2) Is the ~6.6% dividend safe given the uncovered 2025 payout? (3) How much state/political risk is embedded post the 2024 CEO fight, especially into Brazil’s 2026 election? (4) Will VBM (copper/nickel) ever de-commoditize the story via an IPO or further partnership, and does the ~$26bn 2024 mark hold? (5) What is the residual dam-liability tail (UK group action, criminal trial) on top of the Brazilian settlements? (6) Why own Vale over Rio/BHP, which carry the same cost advantage without the Brazil/governance/dam discount?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Below mid-cycle. FY2025 GAAP earnings ($2.35bn) are charge-depressed (impairment + tax + FX) and understate operating power; Q1’26 run-rate (~$6–8bn normalized) is mid-cycle-ish but well below the 2021 peak ($22.4bn). EBITDA ($14.4bn) is ~40% of the 2021 peak ($33.9bn). Interpretation: earnings are in the lower-middle of the cycle on price, near the top of the recovery on the stock.

Driven by external environment or internal actions? Overwhelmingly external — the iron-ore price (set by Chinese steel demand and seaborne supply) drives ~70%+ of the outcome. Internal actions (cost-out, volume recovery, copper growth, buybacks) matter at the margin but cannot overcome the price.

How stable are revenues? Unstable — revenue swung $39.5bn → $54.5bn → $38.4bn over five years almost entirely on price. There is no recurring/contracted revenue cushion; volumes are sold at index-linked/spot prices.

Outlook for products/services? Iron ore: flat-to-down on price (peak China steel + Simandou supply), modest volume growth. Copper: structurally positive (electrification demand). Nickel: structurally negative (Indonesian oversupply). Pellets/high-grade: a slow “green steel” optionality.

How big is the market — growing/shrinking, domestic/international? Seaborne iron ore (~1.5bn tonnes/yr) is mature-to-plateauing, ~70%+ China-destined; the demand center is shrinking at the margin. Copper is a growing global market. Vale is overwhelmingly an exporter (Brazil → China/Asia/Europe).

Business Quality & Competitive Moat

Is the industry getting more or less competitive? More — on the supply side, Simandou adds ~120 Mtpa of high-grade ore into a plateauing-demand market, ending the 2016–24 discipline era (Marathon late-cycle).

How profitable is the business (ROIC, ROE)? Through-cycle ROIC mid-teens (47% at 2021 peak → 8.8% charge-depressed 2025; ~17–20% in 2023–24). ROE 3.8% (2025, depressed) to 36% (2021). Above cost of capital through-cycle, far above at the top — the empirical proof of the cost moat.

How profitable is the industry — competitors, barriers to entry? Big-4 oligopoly (~70%+ seaborne); enormous capital/infrastructure barriers (a new mine+rail+port is a multi-decade, tens-of-billions project). But the product is a price-taking commodity, so the profit pool is set by the marginal tonne, not by the structure.

Can the business be easily understood? Yes — it mines ore and sells it at a market price; the variables are price (China), cost (BRL, grade, freight), and volume.

Undermined by foreign low-cost labor? N/A in the labor sense; the relevant analog is foreign low-cost ore — Simandou (Guinea) and, in nickel, Indonesian HPAL. Yes, foreign low-cost resources are the competitive threat.

Do brands matter? No. There is no brand/demand moat; steelmakers buy on price and grade. The only “differentiation” is the Carajás high-grade quality premium.

Nature of competition / switching costs? Pure cost competition; effectively zero customer switching costs (fungible commodity within grade bands).

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The Carajás ore body’s economic value (decades of low-cost reserves) and the integrated rail/port logistics network are carried at cost/depreciated and are worth far more than book in a normal-price world. The ~90% VBM stake is marked privately at ~$26bn EV (2024 Manara deal) vs its carrying value.

Off-balance-sheet liabilities? Largely on balance sheet now via “expanded net debt,” which adds the Brumadinho/Samarco and de-characterization provisions to conventional net debt ($17.8bn expanded vs $13.6bn conventional, Q1’26). The genuinely off-balance-sheet tail is the UK Mariana group action (Vale ~50% of BHP’s exposure, unquantified) and the criminal-trial outcome.

How conservative is the accounting? Reasonable-to-conservative currently — Vale took a large ~$4.6bn impairment (nickel) in 2025 rather than deferring, and provisions the dam liabilities transparently. IFRS reporting in USD. The 57% 2025 effective tax rate is a one-off distortion.

How CapEx-hungry? Heavy. Capex ~$6.0bn (2025), guiding ~$6.5bn — sustaining capex plus copper growth and the S11D expansion. This is a capital-intensive heavy-industry business, not a capital-light compounder.

Capital Allocation & Management

How much FCF, and how is it used? FY2025 FCF ~$2.8bn (trough; mid-cycle $7bn+). Uses: dividends + interest-on-capital first (~$3.56bn in 2025, exceeding FCF — debt-funded gap), then buybacks (~16% share-count reduction since 2020, at sensible prices), then growth capex. Philosophy: a minimum policy of ~30% of (EBITDA − sustaining capex) plus extraordinary distributions when leverage is below the ~$15bn expanded-net-debt midpoint.

Significant acquisitions recently? Net disposals, sensibly: sold 10% of VBM to Manara Minerals (~$2.5bn, ~$26bn implied EV, 2023–24); progressively sold down PT Vale Indonesia (to ~34%, regulatory-driven). No value-destroying acquisitions recently.

Buying back shares? Yes — an active program (up to 120M sh / ~3%, 18-month), a tranche finalized May-2026; ~5M shares in Q1’26. Buybacks have been at cyclically sensible ($11–15) prices.

Issuing large amounts of new shares to insiders? No — SBC is immaterial; the share count is falling.

Compensation policy? PSU plan now includes a ROIC hurdle (~25% weight from 2024) + TSR + ESG, plus a CEO equity-holding requirement (36x fixed fee). Improved alignment to returns. Insider ownership negligible (FPI norm).

Motivations of management? CEO Gustavo Pimenta (internal ex-CFO, since Oct-2024) and CFO Marcelo Bacci are operations/finance-focused; the stated strategy is operational excellence + disciplined capital allocation + copper/iron-ore growth. Interpretation: aligned with shareholders on returns, but the board sits in a contestable governance arena (no controlling holder, Previ influence, demonstrated state pressure), which is the real management risk.

Valuation & Market Data

ADR, MLP, or K-1? ADR — each VALE ADR = one Vale ordinary share (a true ADR, no ratio mismatch). No K-1; standard 1099 dividend reporting (with Brazilian withholding; interest-on-capital is taxed). Not an MLP.

Dividend policy? Variable semi-annual dividend + interest-on-capital; ~6.6% trailing yield; ~30%-of-(EBITDA−sustaining-capex) minimum plus extraordinary distributions. Caveat: not fully FCF-covered at trough prices.

How profitable is the business? Very, at mid-to-high cycle (35%+ EBITDA margin, mid-teens+ ROIC); merely break-even-plus at the bottom. The cost position sets a high margin floor.

Net income diverging from cash from operations? Yes, and favorably in 2025 — OCF ($8.8bn) far exceeded GAAP net income ($2.35bn) because of the ~$4.6bn non-cash impairment and deferred-tax items. A QoE green flag (cash > accounting earnings).

Risks & Downside

What would cause the stock to decline? A falling iron-ore price (China demand weakness, Simandou oversupply) — the dominant driver; a stronger US dollar; a dam-litigation adverse development; state interference (windfall tax, board politics); a dividend cut; a new operational incident.

Risk of catastrophic loss? Low financially (hard assets, IG balance sheet, lowest-cost position keep Vale solvent through downturns), but the tail risk that would permanently impair the equity is another tailings-dam disaster — precedented (Mariana, Brumadinho) and being slowly de-risked via de-characterization.

Chance of a total loss? Very low. Bankruptcy is implausible absent a second catastrophic disaster; the cost position and asset base provide deep downside support.

Recent News & Events

Has the business environment changed recently? Yes, in two opposing directions: (1) Cleared overhangs — the Mariana settlement (Oct-2024) and CEO succession (Oct-2024) removed two big discounts and drove a +80% re-rating. (2) New headwinds — Simandou first ore (Nov-2025), China peak-steel, and iron ore softening toward $100 introduce a structural supply-cycle risk the stock has not fully discounted.

Significant acquisitions? None recently (net disposals — VBM stake, PT Vale Indonesia).

Change in accounting policies? No material change; the 2025 nickel impairment and 57% tax rate are events, not policy changes.

Recent changes — new markets, facilities, management? New CEO (Pimenta, Oct-2024) and VBM CEO (Shaun Usmar); Serra Sul +20 (S11D +20Mt) starting H2’26; record copper output; a ~$1.5bn upward 2026 FCF revision (May-2026); JPMorgan Overweight/$21 PT (Jun-2026); Cosan exited its ~4% stake (Jan-2025).

APPENDIX B — Source Appendix — Vale S.A. (NYSE: VALE)

Primary sources prioritized. Accessed 2026-06-19 unless noted. All material facts trace to public primary sources.

Primary filings (SEC EDGAR, CIK 0000917851)

• Form 20-F, FY2025 (filed 2026-03-27) — annual report (IFRS, USD): segment revenue mix, share count (4,268,780,141 common + 12 golden), dam provisions (Brumadinho/Samarco/de-characterization), risk factors, governance/golden-share, royalty (CFEM). https://www.sec.gov/Archives/edgar/data/917851/000129281426001844/valeform20f_2025.htm
• Form 20-F, FY2021–FY2024 (filed 2022-04-14, 2023-04-13, 2024-04-19, 2025-03-28) — multi-year history.
• 6-K, Q1 2026 results (reported 2026-04-29; 6-Ks filed Apr-2026) — revenue $9.26bn, adj EBITDA $3.83bn, attributable NI $1.89bn, C1 $23.6/t, copper 102kt, expanded net debt $17.8bn, Brumadinho ~81% / Samarco reparation progress. https://www.sec.gov/Archives/edgar/data/917851/000129281426002648/vale20260428_6k.htm
• 6-K, FY2025 production & results (Jan/Feb 2026) — 336 Mt iron ore, 335–345 Mt 2026 guide, nickel impairment. https://www.sec.gov/Archives/edgar/data/917851/000129281426000189/vale20260127_6k.htm
• 6-K, buyback program (2025-02-19) — up to 120M shares / ~3%, 18-month program. https://www.sec.gov/Archives/edgar/data/917851/000129281425000458/vale20250219_6k1.htm

Earnings call

• Vale Q1 2026 earnings call transcript (2026-04-29) — CEO Gustavo Pimenta, CFO Marcelo Bacci, VBM CEO Shaun Usmar: C1 cost, expanded net debt target ($10–20bn, ~$15bn midpoint), $2.7bn Q1 distributions + ~5M sh buyback, “increasingly confident on extraordinary dividends,” Serra Sul +20 86% complete, copper/nickel records. (public earnings-call transcript).

Quantitative data (reconciled to filings)

• Public financial data aggregators (company filings reconciled) — income statement, balance sheet, cash flow, profitability ratios, enterprise value (6-year annual): revenue/EBITDA/net-income series, ROIC/ROE trajectory, net debt, FCF, dividends. (Third-party aggregated; reconciled to 20-F.)
• Daily price/OHLCV history (VALE) — 5-year OHLCV, EMAs, beta: price event map levels ($22.94 peak Jul-2021; $8.26 trough Apr-2025; $15.42 Jun-18-2026), 21/50/200-EMA, beta 0.87.
• Quantitative factor/risk model — factor loadings (Industry-Mining +1.24, Country-Brazil +0.59, USDollar −0.53, Quality −0.20, GoldPrice +0.38), leaderboard (y1 +80.7%/Sharpe 2.50/maxDD −19.9%), stock-info (rs_12m +79.8, rs_peak −13.5), related stocks (RIO 0.93, BHP 0.89, FCX 0.84).

Industry, disasters, governance, news (public secondary)

• Mariana/Samarco definitive settlement (~R$170bn, Oct-25-2024): Vale press release; NS Energy. https://vale.com/w/definitive-settlement-mariana
• English High Court, Município de Mariana v BHP (Nov-14-2025) — BHP liable, Vale ~50% contribution: Sidley insights; Pogust Goodhead. https://www.sidley.com/en/insights/newsupdates/2025/12/english-high-court-finds-bhp-liable-in-landmark
• Brumadinho settlement (R$37.7bn, Feb-2021) and criminal trial (began Feb-2026): Reuters/Nasdaq; Business & Human Rights Resource Centre.
• CEO succession / Lula pressure / Gustavo Pimenta (Oct-2024): Bloomberg/BNN; Benzinga; Vale PR.
• Simandou first ore (Nov-2025) and ramp to ~120 Mtpa: Maritime Executive; Ecofin Agency; MINING.com (“$100 floor at risk”). https://maritime-executive.com/article/first-cargo-of-iron-ore-from-giant-simandou-mine-arrives-in-china
• Iron-ore price (~$101–107/t mid-2026, softening): IndexBox (June 2026); Stockhead (2021 Evergrande crash). https://www.indexbox.io/blog/iron-ore-prices-decline-in-late-may-early-june-2026/
• Vale Base Metals — Manara Minerals 10% sale (~$2.5bn, ~$26bn EV): MINING.com (Jul-2023); Vale PR (2024). https://www.mining.com/vales-energy-transition-metals-business-attracts-3-4-billion-investment/
• VBM “IPO-ready by 2027” / Usmar copper strategy: Bloomberg/MINING.com (Jun-2025). https://www.mining.com/web/vale-base-metals-boss-says-unit-aims-to-be-ipo-ready-by-2027/
• PT Vale Indonesia / MIND ID majority (2020, 2024): Mining-Technology; MIND ID.
• 2025 nickel impairment (~$3.5bn) / FY2025 net income: Reuters via Investing.com (Feb-2026); MINING.com. https://www.mining.com/web/vale-posts-3-8-billion-q4-loss-on-nickel-assets-impairment/
• JPMorgan Overweight PT $21 (Jun-12-2026); BofA 2026 top pick; consensus aggregates: Yahoo/Insider Monkey; TipRanks; MarketBeat; public.com.
• 2026 FCF outlook raised ~$1.5bn (May-12-2026): Globe & Mail. https://www.theglobeandmail.com/investing/markets/stocks/VALE/pressreleases/1905437/vale-lifts-2026-cash-flow-outlook-on-iron-ore-rally-and-details-nickel-sensitivities/
• Cosan exits ~4.05% Vale stake (Jan-16-2025): US News/Reuters; Investing.com.
• CFEM royalty (3.5% iron ore): Lexology; Mondaq.

Frameworks applied

• Greenwald & Kahn, Competition Demystified — cost/scale advantage taxonomy, share-stability and ROIC tests .
• Marathon / Chancellor, Capital Returns — supply-side capital-cycle analysis (Simandou as the late-cycle supply inflection).