A record half — spent selling the future.
On the face of it, BHP’s HY26 was one of the strongest results of Mike Henry’s tenure: underlying EBITDA of US$15.5bn, up 25%, at a 58% margin; attributable profit of US$6.2bn; and operating cash flow of US$9.4bn. For the first time in the company’s history, copper delivered more than half of group EBITDA, at a 66% segment margin, while Western Australian iron ore printed a record half at 62%. Net debt sat at a comfortable 0.5× EBITDA, return on capital reached 23.6%, and the interim dividend held at 73 US cents.
And yet, against that backdrop, BHP announced three disposals that together pull US$6.3bn of cash forward — with a stated path to US$10bn. The headline act was a Wheaton streaming deal that sold BHP’s attributable silver at Antamina for US$4.3bn upfront: the largest precious-metals streaming transaction ever struck. Alongside it, the Carajás divestment and a US$2bn inland-power deal on the iron-ore business. Weeks earlier, the Board had named Brandon Craig — the architect of much of the growth pipeline — to succeed Henry from July. When a company this strong is pre-selling its cash flows, the interesting question is not the framing. It is what the numbers are telling you about the risk it now carries.
The silver deal was too good — which is exactly the warning.
The most revealing line in the HY26 deck is footnoted, not headlined: the upfront proceeds from the silver stream approach the consensus NAV of BHP’s entire 33.75% stake in Antamina. Read that again. BHP received, in cash, roughly what the market values the whole interest at — by selling only the silver byproduct, and keeping the copper, zinc, lead and molybdenum. Either Wheaton overpaid and BHP captured a windfall, or the sell-side has been quietly under-marking BHP’s non-operated joint ventures for years. Probably both.
But a windfall at execution is not the same as a good trade over time. The stream has no buyback clause and runs for the life of the mine; BHP now keeps just 20% of the spot silver price on every ounce delivered. In other words, it has pre-sold 80% of the upside on precisely the commodity Wheaton is betting will rise. And the accounting is its own tell: management states the US$4.3bn is “not expected to increase reported debt levels.” In substance, this is leverage. In form, it is not.
You do not sell your best byproduct at the top of its cycle unless you need the cash somewhere else — and BHP’s capex tells you exactly where.
The growth bill is larger and longer than the headline. Cumulative capex across FY26–30 runs to roughly US$52bn, near US$58bn once non-operated JVs are counted, and the execution record has cracked. Jansen Stage 1 was sanctioned at US$5.7bn and now stands at US$8.4bn — a 47% overrun that, on BHP’s own admission, has dragged its project-delivery variance to about 25%, against roughly 3% on everything else it builds.
That is why the disposals matter. Over the past decade BHP returned more than US$110bn to shareholders — around 70% of its market capitalisation — and the market has priced that cadence in. Run the affordability math through the capex peak and the distributable cash left after sustaining capital, growth, JV calls, tax and Samarco falls to roughly US$3–5bn a year, below both the decade’s pace and the implied minimum dividend. It holds at spot; it does not hold in the downside.
