Vale Is Trading Like Iron Ore Stays Below $80 Forever

The 8.2x forward earnings multiple and a target price of $17.27 from sell-side consensus encode an implicit iron ore price assumption somewhere in the $80-90 per tonne range for the next several years. That assumption is at odds with the supply outlook. Simandou ramps in Guinea over 2026-2028 and adds roughly 60 million tonnes of high-grade supply at the margin. Strip that incremental supply out and the seaborne market is short, not long.

The Chinese rebar market is weaker than 2021, agreed. But global crude steel production was up 1.6% in the most recent quarter on the back of Indian and ASEAN demand. India alone is on track to add roughly 30 million tonnes of new steel capacity over the next 18 months, and that capacity is heavily skewed toward high-grade pellet feed. Vale's Carajás system produces some of the highest-grade iron ore on the seaborne market at 65%+ Fe content; that grade premium is widening, not compressing.

There is also a question of how much extra discount the consensus has baked in for the Brumadinho settlement obligations. The remaining Mariana and Brumadinho cash outflows are largely book-provisioned and the schedule is known. Pricing them as a perpetual drag rather than a finite obligation has been the market's mistake for two years running.

Historically, when iron ore producers have traded at single-digit forward earnings multiples with cost positions in the lowest decile, the subsequent 24-month total return has been positive in nine of the last eleven cycles. Freeport-McMoRan in 2020 is the textbook case for the read-through; it traded at 7x forward earnings with copper at $2.80 per pound and re-rated 110% over 18 months as the supply deficit began pricing in.

Vale's copper production is on track for roughly 380-400 thousand tonnes in 2025, with Salobo III contributing meaningfully and the Hu'u project moving toward final investment decision. By 2030, the company has guided to copper production approaching 700 thousand tonnes, putting it in the conversation with Antofagasta, Southern Copper, and Teck for top-tier copper exposure.

Apply a peer copper multiple to that 2030 output stream. At $9,500 per tonne copper and a 35% EBITDA margin, the implied EBITDA from the copper book runs at roughly $2.3 billion. Apply a 6.5x EV/EBITDA, which is conservative against the pure-play copper peer average of 7-8x, and the embedded business is worth somewhere between $14 and $18 billion. The market is currently valuing the entire iron ore franchise at roughly $57 billion enterprise value and giving the copper book a value of approximately zero. That is a striking analytical error.

Look, the copper expansion is not without execution risk. Hu'u has had permitting delays. Salobo III ramp is a year behind original schedule. The base nickel business has had operational and pricing challenges. None of these are fatal. They explain why the copper book trades at a discount; they don't explain why it trades at zero.

The Northern System produces iron ore at a C1 cash cost in the low to mid $20s per tonne. Even at $80 iron ore, that is a 60-70% cash margin. The high-cost Chinese seaborne marginal supplier sits closer to $90 break-even. Mathematically, this means Vale's earnings are far less sensitive to iron ore price than the broader market complex; it is the high-cost producers who get squeezed first when prices fall, and Vale is the lowest-cost major producer at scale.

The BRL has been a tailwind in 2024-2025 as the real weakened against the dollar, supporting USD-denominated unit costs. Even on a normalised currency basis, Vale's all-in sustaining cost runs $35-40 per tonne, comfortably below the seaborne 90th percentile cost curve. The cost position is not eroding. If anything, the new shipping fleet and the Northern System logistics improvements have widened the cost gap to the marginal Chinese-sourced inland producer over the last three years.

Capex is, frankly, restrained. Capital expenditure in 2025 came in around $5.7 billion, a 30% reduction from peak years and a level that comfortably supports the dividend without consuming the entire cash generation. The question is not whether the dividend is safe; the question is what coverage ratio justifies the current price. Even at $80 iron ore, the dividend coverage stays above 1.5x on our model.

Normalise the cycle and the picture sharpens. Across the 2021-2025 window, average operating income was roughly $16 billion and average free cash flow was approximately $9 billion. Apply a 6x EV/EBIT multiple to the normalised earnings power and the implied enterprise value is in the $96 billion range. Today's enterprise value sits at $87.7 billion. That is a 10% gap on a midcycle anchor, but the asymmetry runs heavily to the upside because 2025 is closer to the bottom of the cycle than the middle.

Go a layer deeper. The book value of the resource base alone, calculated from reserves at a long-run iron ore price of $90 and copper at $4.50 per pound, exceeds the current market cap by a meaningful margin. Vale's net debt position of approximately $13 billion is comfortably serviced by the operating cash flow line and represents under 1.0x EBITDA. There is no balance sheet stress here.

The trailing dividend yield on the ADR sits in the 7-9% range depending on the basis used. The forward yield, after the 2025 distribution recalibration, lands closer to 5-6% sustainable. Either way, the income coupon while waiting for the re-rating is more than competitive against the energy and basic materials peer average. Over the past three years, Vale has bought back roughly 11% of shares outstanding at average prices below current levels. That is capital allocation discipline.

Look at the analyst rating distribution: 8 strong buys, 9 holds, zero sells. That is not a name where the sell-side consensus is bearish. It is a name where the share price has compressed despite a constructive analytical consensus, which often signals positioning unwind rather than a fundamental view change.

Look at the historical analogues. BHP traded at 0.9x book in late 2015 with iron ore at $40 per tonne; the subsequent five-year total return was 240%. Rio Tinto traded at 1.0x book in early 2016 with the same backdrop; the subsequent five-year total return was 215%. Vale itself traded at 0.9x book in mid-2020 during the early pandemic; the subsequent 18-month total return was 180%. Across three complete cycles, the pattern is the same: market over-discounts the trough, miners earn through the cycle, and the re-rating delivers outsized returns.

The specific sequencing matters. In each of those analogue cases, the re-rating did not begin with a commodity price recovery. It began with a stabilisation in the price decline rate. The market needs evidence that the bottom is in, not that the cycle has turned upward. That is a low bar. Iron ore has spent the last 14 months trading in a $90-110 range. The variance has compressed. The bottom signal is forming.

The other lesson from the analogues is that capital discipline matters more than commodity price in determining the multiple expansion. Vale's capex run-rate has come down 30% from peak. Net debt is under 1.0x EBITDA. Buyback execution has been disciplined, with roughly 11% of shares retired over three years at average prices below current levels. These are the operational signatures that precede the multiple re-rating, not follow it.

The 50-day moving average sits at $16.18 and the 200-day at $13.07. The cross is constructive. The 52-week range of $8.24 to $17.94 captures the volatility, but the volume profile shows institutional accumulation in the $13-15 zone over the last six months. That is the buy zone the market has chosen. We agree with it.