THERE are few words in South African economic policy more seductive than “beneficiation”. It has everything: moral uplift, industrial ambition, a pleasingly muscular rhythm and the promise that somewhere, just beyond the next cabinet lekgotla, a lump of ore will be persuaded to become a dishwasher, a turbine blade or perhaps a small electric car by force of patriotic will.
It is the sort of word that makes policy documents stand up straighter. It says: “Why should we export rocks when we can export things?” Why must South Africa forever be the quarry and never the machine shop?
The recent sale by South32 of most of its aluminium assets to Alcoa offers a wonderfully inconvenient case study. South32 is selling a large chunk of its aluminium value chain to the US aluminium giant: Worsley Alumina in Western Australia, bauxite and alumina interests in Brazil, and, most importantly from a South African point of view, Hillside Aluminium in Richards Bay.
It’s a thumping transaction, and I think a pretty good one for all sides from a corporate finance point of view. When I first saw the deal, my immediate thought was, despite the public relations gloss, it’s essentially yet another divestment by an international mining company. But on the face of it at least, the advantages for all sides are apparent.
For South32, the effect is that it tilts towards higher-margin base metals, especially copper, zinc, lead and silver. It effectively removes a large capital-intensive, lower-growth part of the portfolio. And then there is the fact that South32 expects to return about $500m to shareholders after completion while keeping balance sheet flexibility for growth or acquisitions. It’s no wonder South32’s shares jumped — investors saw a cleaner company, a decent price and the prospect of juicy cash payback.
For Alcoa, it’s a bit more complicated, but what is undisputed is that it adds substantial heft. Alcoa’s pro forma 2025 numbers suggest a 53% increase in alumina production, a 37% increase in aluminium production, and a 45% increase in adjusted ebitda. So why did the share price go down after the deal was announced? One, because the company is paying partly in shares, and two, because it racks up some debt.
But on the numbers, it’s a pretty good deal for Alcoa. The assets are being bought for implied enterprise value/ebitda of 5.2 times to 6.1 times, depending on whether the contingent value right pays out. Alcoa compares that with its five-year average EV/ebitda multiple of 6.3, which makes the deal look roughly “through-cycle”, not a fire sale.
But there are two big qualifications. First, 2025 was a good year for alumina: South32’s alumina earnings were lifted by a 45% rise in realised prices, so the sale multiple is being calculated on a healthy earnings year, not obviously a trough. Second, the assets are not all jewels. Alcoa’s presentation shows Worsley-Boddington as strong, Brazil as strong, but Hillside as third quartile on the 2025 cost curve.
Presumably, Alcoa imagines it could do something about that. And from Hillside’s point of view, there are obvious advantages in being locked into the sales and distribution system of one of the biggest companies in the sector. Alcoa has pencilled in $900m of synergies, which is impressive. But a lot depends on discussions with Eskom about electricity price and availability.
Eskom love/hate
Eskom both loves and hates Hillside, I would imagine. It is a huge consumer that actually pays its bills, which makes a change for Eskom. But South African power costs have risen roughly tenfold since 2008, and Hillside’s economics depend on a negotiated electricity price that is far below the ordinary large-user tariff. Thanks to the Eskom-Hillside agreement published by Open Secrets, we know the formula rather better than we used to: Meridian Economics estimates that the deal gives Hillside about a 50% discount to the standard Megaflex tariff over the 10-year term.
Internationally, it is less obviously scandalous because aluminium smelting survives only on long-term, competitive power. The fact that South32 put Mozal on care and maintenance after failing to secure sufficient affordable power tells its own story.
And this is where the parable on beneficiation comes in. Australia has mountains of bauxite and serious alumina-refining capacity. South Africa, by contrast, has no meaningful bauxite and no large alumina-refining industry. Yet for decades it had a globally significant aluminium smelter. The reason is that Hillside was never really the beneficiation of bauxite; it was the beneficiation of electricity.
South Africa imported alumina and turned it into aluminium because, at the time, the country had cheap, abundant coal-fired power, a deepwater port at Richards Bay and a state power utility willing to sign industrial electricity contracts that made the whole thing work.
It demonstrates that South Africa can and does beneficiate, but not if beneficiation is a punishment imposed on miners for being insufficiently patriotic, which is effectively what the trade, industry & competition department has suggested. It is a commercial outcome produced by competitive inputs. Cheap and reliable electricity. Efficient ports. Functional rail. Clear laws. Fast permits. Tax stability. Skills. Technology. Sensible incentives. Market access.
South Africa still seems a little vague on the difference between encouraging a value chain and mugging it at the border. This deal is not a simple story of South Africa losing another industrial trophy, but it is a reminder that industrial assets live inside systems.
This article first appeared in the Financial Mail.
