Dr Duarte da Silva: Sibanye's Burnstone decision

Burnstone is what a South African gold board looks like when it decides to build. On 1 September 2026 Sibanye-Stillwater put about R6.2 billion behind a mothballed, once-liquidated, twice-deferred asset and called it a twenty-five-year mine at a 36.1 per cent internal rate of return. Kloof, for a generation among the richest reefs in the basin, got one sentence in the same results about assessing what might lie beyond 2026. This paper reads both decisions in full, because the second is only defensible once the first has been..By Dr Duarte F da Silva.Kloof carries a single year of declared life, fixed on 17 February 2026 at a reserve price within a fifth of the highest ever recorded, three weeks after gold printed its all-time high. That declaration — a Mineral Reserve cut from 1.6 million ounces to 0.1 million — has not been revised. This paper does not re-argue it. It tests it against a project the same board approved on the same day, and asks what the comparison changes and what it does not. The case for Burnstone is put at its strongest, because it is strong. Strength on one asset is not credit against another, and the two questions are kept apart: what Burnstone proves about this board's capacity to build, and what it changes about the case for holding Kloof rather than flooding it..EXHIBIT A · THE SAME FILING, TWO ANSWERSI. One document, read wholeSibanye-Stillwater's results for the six months to 30 June 2026, published on 1 September, gave Kloof one sentence. The supportive gold-price environment, the company said, “provides an opportunity to assess the potential economic extraction of additional reserves at Kloof beyond 2026, including through the support of appropriately priced hedging mechanisms.” On the call, management put a number on it: at about R2.4 million a kilogram, Kloof could be looked at for a life extension to about 2029, possibly underwritten by a hedge book on a specific block of ground. Then came the qualifiers — nothing committed, under assessment, subject to returns and affordability. No reserve was restated. No capital was allocated. The filing confirmed no material change to the Mineral Reserves declared six months earlier: Kloof's hundred thousand ounces and its single year of life stood exactly as fixed on 17 February 2026.The same results carried a second announcement. Burnstone, the shallow gold project near Balfour that Sibanye has owned in one form or another since 2014, had “recently received a positive investment decision by the board to proceed.” Not an opportunity to assess — a decision to proceed, backed by a restated feasibility study, an internal assurance review, a disclosed net present value and a disclosed internal rate of return. This paper asks the board to bring that rigour to Kloof. The proof that it can is in the same document, on the same day, at a different address.Not an opportunity to assess — a decision to proceed. The proof that this board can run the test is in the same document, on the same day, at a different address.EXHIBIT B · THE CREDIT — WHAT WAS APPROVEDII. What was approvedBurnstone carries, at 31 December 2025, Mineral Reserves of 2.7 million ounces — 19.9 million tonnes at 4.2 grams a tonne — and Mineral Resources the company puts at 8.9 million ounces, on the Kimberley Reef at an average depth of about 550 metres and a maximum of 1,050. The restated feasibility study, in 2026 real terms, returns a net present value of about R19.2 billion at a 10 per cent discount rate and an internal rate of return of about 36.1 per cent — more than double the group's 15 per cent hurdle and second in the current pipeline only to the K4 platinum project's 99 per cent. Total project capital is about R6.2 billion: R3.5 billion of infrastructure capital over six years, R2.5 billion of pre-production capital to 2028, and R98 million approved for 2026, mostly set-up and recruitment.At steady state the mine is to produce about 130,000 ounces (four tonnes) a year for twenty-five years at an all-in sustaining cost of R872,000 a kilogram. At the half-year's average rate of R16.41 to the dollar that is about US$1,710 an ounce. Against spot gold of US$4,480 at the close on 4 September 2026, and an average received price of US$4,597 over the half, the mine carries about US$2,770 of headroom on every ounce before it turns marginal — a margin structure closer to Mponeng's or DRDGold's than to Kloof's. Mining starts in 2027, ore is stockpiled, and the plant begins operating in the first quarter of 2029. The project is to employ about 2,500 people at steady state, in a sector whose headcount fell from just under 120,000 in 2014 to a little over 90,000 in 2024.R872,000 a kilogram against gold at R2.4 million: about US$2,770 of headroom on every ounce, before a tonne has been hoisted.EXHIBIT C · ELEVEN YEARS, THREE OWNERS, TWO NEAR-DEATHSIII. The road it took to get hereThe history is the credit, and it is longer and harder than a summary allows. Burnstone was built by the Canadian junior Great Basin Gold as a mechanised long-hole stoping operation with a plant commissioned in January 2011 — one industry visitor in its opening weeks called it state-of-the-art, without irony. The engineering did not fail. The financing did. Great Basin had raised C$149.5 million in equity in April 2007, most of it spent buying out its partner at Hollister in Nevada, and a further C$126.5 million in convertible debentures in November 2009 to finish Burnstone. It was not enough. By 2012 the ramp-up had hit unforeseen faulting and water problems, guidance for the year was cut from 90,000–100,000 ounces to 30,000, and a strategic review committee convened on 15 August recommended, within a month, suspending everything.Great Basin suspended Burnstone on 11 September 2012 rather than fund it to break-even, which its own numbers put at May 2013. It owed some US$150 million on a secured Burnstone facility, about US$126 million in unsecured convertibles and other debt besides — close to US$300 million in all. It sought creditor protection in Canada in the third week of September and was delisted from Toronto on 25 October, five weeks later. The South African operating subsidiary, Southgold Exploration, went into business rescue under Chapter 6 of the Companies Act the same year.The rescue nearly did not happen, and it nearly ruined the rescuer. Witwatersrand Consolidated Gold Resources — Wits Gold, a far smaller listed miner run by Philip Kotze — was named sole preferred bidder, and Southgold's creditors, Credit Suisse and Standard Chartered among them, approved the sale on 11 July 2013. The terms tell the story: a US$7.25 million down payment, funded from Southgold's own free cash flow, against liabilities that had stood at US$170 million. Kotze told the trade press what failure would mean — “If we fail at Burnstone, then that's the end of Wits Gold” — and Miningmx ran it as the headline. It did not fail. Sibanye Gold, midway through Neal Froneman's sweep of the basin's unwanted assets, bought Wits Gold outright for R407 million, effective 14 April 2014; the Burnstone leg closed in the first week of July 2014 when Southgold's twenty-two-month business rescue ended. Mining Weekly reported in 2021 that Sibanye then bought the project itself from Wits Gold for a nominal R1 in 2015; this paper has not traced that internal transfer to a filing and states it on Mining Weekly's account alone. A feasibility study followed in 2015, development began in 2016, and paused again in 2018.The board first approved continued construction in the first quarter of 2021, on a case a fraction of today's: R2.8 billion of project capital plus R1.5 billion pre-production, a 20 per cent IRR and an NPV of R1.3 billion at R800,000 a kilogram in 2022 real terms. It did not hold. In November 2023, with PGM and nickel prices falling, Sibanye deferred Burnstone “to preserve cash in terms of the Group's Capital allocation framework” — the year it booked R47.5 billion of impairments against a portfolio that then leaned far more heavily on platinum than gold. Deferring discretionary growth under that pressure was prudent, and this paper does not dispute it. The project stayed deferred through 2025, the company stating that it carried no technical or strategic problems, only an investment decision that had to weigh the renewal of its mining right and possible changes to mining law — a specific, still-live uncertainty taken up in Section IX.In February 2026, asked about capital intensity at the results presentation, chief executive Richard Stewart told RMB Morgan Stanley's Chris Nicholson that “the plant's largely built,” with a decision expected in the first half. At the capital markets day on 23 June, with pre-production capital then at R2.7 billion, he told the room that “Burnstone will require renewal” of its mining right, that “there are a lot of rumours around what might come out in terms of renewals of new mining licences,” and that a project committing capital over “five to six years” needed regulatory certainty before his board would sign. The decision came two months later than his guidance, on 1 September — pre-production costed at R2.5 billion inside a total of R6.2 billion — at the fifth juncture since 2015 at which this project has stood before a board asking whether to proceed..EXHIBIT D · THE RAND THAT WENT THE OTHER WAYIV. Two assets, one nominal figureBlyvooruitzicht changed hands in 2012 on terms engineered to be small. Village Main Reef bought 74 per cent of the mine from DRDGold for R150 million of loan claims, settled in Village shares, and a single rand for the shares themselves; a second leg was never completed before the whole of it went into liquidation the following August. Burnstone, on Mining Weekly's account, moved inside Sibanye for the same rand three years later. A rand on the equity is not a price. It is a seller's admission that an asset's value has fallen to the level of its liabilities — and on these two mines that admission has been wrong both times.At Blyvoor the rand led to a walk-away within eighteen months: liquidation, a workforce stranded between employment and retrenchment, a village of six thousand that needed a court order to keep its taps running. At Burnstone the rand led, over eleven slower years, to a company that kept paying — a feasibility study, a false start, a pause, a first approval, a deferral, and a restated case with a nineteen-billion-rand NPV attached. The road ran through a decade of the owner's own hesitation before it ran toward the number now on the table. But it ran toward more capital, not less, and it ended at a signed decision rather than a liquidator's notice. Both mines were marked down to a rand on the day their owners stopped believing in them at the price then on offer. Only one owner came back, and it took most of a decade.Both mines were marked down to a rand on the day their owners stopped believing in them. Only one owner came back — and it took most of a decade.EXHIBIT E · THE ARITHMETICV. What Burnstone would cost Kloof, if Kloof got the same testThe one place a boardroom compares two projects is capital per ounce. Burnstone's R6.2 billion against 2.7 million reserve ounces is a little under R2,300 an ounce of development capital — about US$140 at R16.41. Apply that intensity, as a scale check rather than a forecast, to Kloof's 4.3 million ounces of Mineral Resource — 3.9 million of them measured and indicated, the ground excluded from reserve rather than mined out — and the equivalent spend is about R10 billion. That is a large number against a single asset and a modest one against this balance sheet. It is less than the R12.5 billion the South African gold segment, DRDGold included, earned in adjusted EBITDA in 2025, close to the R9.0 billion the same segment earned in the six months just reported, comparable to the R7.9 billion Harmony committed in February 2024 to going deeper at Mponeng, and less than double what Sibanye has just approved for Burnstone.Run the check the other way. Kloof's surviving reserve — the hundred thousand ounces left after the February writedown — would at Burnstone's intensity cost about R230 million to bring to production, some two and a third times the R98 million already released for Burnstone's 2026 set-up alone. Kloof's ground and Burnstone's are not interchangeable engineering problems, and nothing here claims a Kloof reassessment would cost precisely what the arithmetic implies. What it establishes is that taking Kloof's question seriously sits well inside the range this board has just shown itself willing to spend on one shallow project. Affordability was never the constraint. The reserve-price convention was.At Burnstone's capital intensity, Kloof's excluded resource is a ten-billion-rand question — against nine billion of gold EBITDA in six months. Affordability was never the constraint.EXHIBIT F · THE FRAMEWORK'S OWN ORDER OF PRIORITYVI. What the company's own rules say should come firstSibanye's capital allocation framework, as set out in the H1 2026 results materials, ranks spending in a stated order. Non-negotiables come first: sustaining capital, ore reserve development capital, and liquidity of at least two months of operating expenses and capex. Only after those are met is cash split in thirds between shareholder returns, debt reduction and growth — the bucket Burnstone, Mount Lyell and K4 all sit in. Burnstone had been on the company's own shortlist of qualifying growth projects since at least 2022, when the annual filing named it alongside K4 and Keliber as one of the few developments clearing the hurdle on low capital intensity and a short lead time to payback. It took four more years, a group-wide impairment, a change of chief executive and two deferrals for that qualification to become a signed cheque.Reassessing Kloof's excluded resource is not, on this framework's terms, a growth decision competing with Burnstone for a place in the queue. It is ore reserve development — the company's own label for the work needed to test whether ground written out of a reserve statement can be written back in — and ore reserve development is named among the non-negotiables that rank ahead of growth capital, not behind it. Nothing in the public record shows capital allocated to that category at Kloof since February. When Stewart set out his refreshed strategy on 29 January 2026 he named South African gold, and Burnstone within it, a strategic priority, in the same breath as naming the South African PGM operations the group's highest priority for growth. Burnstone was never Sibanye's first choice of where to spend; it was the gold project that cleared the bar while a platinum project cleared it faster. Kloof was not entered in the competition at all.None of this implies bad faith. A company managing three metals on four continents through the worst PGM downturn in a decade had reason to sequence its non-negotiables carefully, and did. But the excuse available in 2023, when cash needed preserving, is not the excuse available now: gross debt is down 18 per cent since December to R32.1 billion, net debt has more than halved to R9.7 billion, an interim dividend of R5.7 billion has been declared, and the Kloof sentence commits none of the ore-reserve-development capital the framework says comes first.Ore reserve development ranks ahead of growth capital in Sibanye's own framework. Kloof is the one asset where that category most obviously applies, and the one where none has been allocated.EXHIBIT G · THE COUNTRY'S LEDGERVII. Two new mines, after fifteen years of noneThe national record has just moved. West Wits Mining's Qala Shallows, in production since October 2025 and pouring since March, was reported on opening as South Africa's first new underground gold mine in fifteen years. Burnstone's approval, eleven months later, is close to a second. Together they are the first sign that the long decline in exploration spending and the country's near-absence from new mine development may be turning at the margin.Neither touches the deep frontier at Kloof and Mponeng. Qala Shallows and Burnstone are both shallow tabular reefs worked far above the Ventersdorp Contact Reef, and both were de-risked before approval — Burnstone by a decade of predevelopment, existing shafts, an existing plant and an existing tailings facility, most of it built and then mothballed. Two shallow mines in under a year is news for a country that had built none in fifteen. It is not evidence that the narrow-reef problem at 3.4 kilometres has become easier, cheaper or less urgent. A company that has assembled a restated feasibility study, an assurance review and a disclosed hurdle-clearing return for one reef, twice in five years, and has not done the equivalent for its richest, has the institutional muscle for the harder job and has not pointed it at Kloof.Two new shallow mines in under a year, in a country that had built none in fifteen. Neither touches the deep problem — and the company that built one of them has the muscle for it.EXHIBIT H · THE NUMBER NOT DISCLOSEDVIII. The disclosure Sibanye didn't make, and the cautions it should hearEvery Mineral Reserve statement Sibanye and its peers have published in this cycle — Kloof, Driefontein, Beatrix, South Deep, Mponeng, Fairview — discloses the gold-price assumption behind it, to the dollar. Burnstone's restated feasibility study, in anything published so far, does not. The company says the case is in “2026 real terms” and leaves the reader to infer the price from the margin. A board that discloses its price to the ounce at Kloof, to explain one year of life, and withholds it at Burnstone, to justify R6.2 billion of conviction, is running two standards of transparency inside one results pack. Transparency is cheap when it justifies caution and dear when it must justify conviction.Two analysts' cautions belong here in full. Peter Major of Modern Corporate Solutions backed the decision — “I have no doubt Burnstone is worth going for now because there is quite a bit of development. The shaft is put in place. I think they'll make money because they bought it pretty cheap” — but warned that the fractured geology has historically made the mine hard to work and that its discontinuous orebody may not carry the tonnages in the plan. Great Basin's 2012 collapse was caused by exactly that disappointment, on this ore. Rene Hochreiter of Noah Capital Markets supplied the sharper number: Burnstone “is not a dripping roast. At current gold prices, it makes a profit, but not much lower.” He put the all-in sustaining margin at spot near 57 per cent on paper and said that if mining costs doubled against the study's assumptions, as they can, the margin could fall to about 13 per cent — inadequate for an investment of this size. There is room in that arithmetic, far more than the US$148 an ounce Kloof carried into the same results. Room is not immunity, and a board capable of misjudging Kloof's downside can misjudge Burnstone's.Kloof's H1 2026 margin, in the same results, was about US$148 an ounce — the thinnest in the South African gold book. Burnstone's disclosed cost structure leaves about US$2,770 against the same spot price.EXHIBIT I · THE UNCERTAINTY THAT DID NOT GO AWAYIX. The licence that expires before the ore doesBurnstone's mining right, granted in 2009, runs to 16 February 2027 — within weeks of the start of mining under the September decision. On 20 May 2025 the Minister of Mineral and Petroleum Resources, Gwede Mantashe, gazetted the Draft Mineral Resources Development Bill, 2025, for comment, proposing changes to the empowerment provisions of the Mineral and Petroleum Resources Development Act. Comment closed on 13 August 2025. At the date of this paper no Bill has been introduced to Parliament. The gap that matters to Burnstone is that the draft, as gazetted, contains no transitional provision recognising empowerment credentials established under the existing order when a right comes up for renewal. Stewart said as much in June: “Historically, Burnstone has been an empowered asset. So in terms of the numbers and the valuations that is the way we continue to look at it.” The company is not worried that Burnstone lacks credentials. It is worried that the credentials it has may not survive the renewal the mine needs before its first full year of production..Read more:.Da Silva’s Gold Papers: Sibanye's plan to flood the richest reef on earth - at best gold price ever.That is narrower and, for a board, more urgent than Hochreiter's general complaint that empowerment rules deter investment. An asset whose empowerment structure was priced as a known cost for over a decade now faces the possibility of a rule change mid-renewal, with a hard date attached and no Bill before Parliament to give the answer. It sits on top of the geological and cost risks, not instead of them. A board that has decided to build around all three deserves credit. A board that has resolved none of the three at Kloof, where no renewal deadline forces the question, has correspondingly less excuse for its silence there.The licence expires on 16 February 2027, weeks after mining starts, under a draft Bill with no transitional clause for the credentials the asset already holds — and no Bill yet before Parliament.EXHIBIT J · WHAT THIS IS NOT PERMISSION FORX. The misreading to rule outThere is an obvious wrong way to read this paper: that a company willing to build Burnstone has earned the benefit of the doubt on Kloof, and that flooding the richer reef is a defensible trade against funding the shallower one. Nothing in the record supports it. Burnstone and Kloof are not a portfolio choice between equivalent assets. They are a demonstration, on one reef, of exactly the capability the board has so far declined to apply to the other — a restated feasibility study, an assurance review, a disclosed rate-of-return test, real capital, a start date. The existence of that capability makes the Kloof sentence harder to excuse, not easier.A company that cannot build is owed some sympathy for what it leaves undone. A company that has just shown, in the same document, that it can — and chooses not to at the one address where the reef is richest and the option is closing fastest — has moved from facing constraints to making a choice. Burnstone does not lower the bar Kloof has to clear. It raises it, by removing the one excuse that might otherwise have covered the decision: that the discipline asked for here is beyond this board's reach.THE VERDICTXI. One board, two reefs, one standard missingThis paper does not ask Sibanye to mine Kloof the way it is building Burnstone. The ground will not allow it, and no amount of capital discipline turns a 3.4-kilometre seismic ribbon into a 550-metre bulk deposit. The ask is narrower, and Burnstone has supplied the proof that it is reasonable: apply to the reef being flooded the same real-terms feasibility work, the same disclosed rate-of-return test, and the same willingness to let a restated case reopen a closed one, that this board has just applied in public to a shallower mine 130 kilometres away.Burnstone earns Sibanye real credit, given here without qualification. A company that let a rand-valued wreck sit through a liquidation, a pause, a stalled approval and a deferral, and funded it anyway, has not given up on South African gold. It has, for the moment, given up on one piece of it — the richest piece, the one furthest along the road to a decision it cannot take back, and the only major decision left in this basin that the board has not yet put to the test it has just proved it knows how to run..*Dr Duarte F da Silva · Capital Markets Strategist · Managing Director, Northbound Processing · Germiston, 2026.Sign up for your early morning brew of the BizNews Insider to keep you up to speed with the content that matters. The newsletter will land in your inbox every morning on weekdays. Register here.Support South Africa's bastion of independent journalism, offering balanced insights on investments, business, and the political economy, by joining BizNews Premium. Register here.If you prefer WhatsApp for updates, sign up to the BizNews channel here.

Dr Duarte da Silva: Sibanye's Burnstone decision

Related Stories

Central banks are dumping the dollar for gold, and SA has left its 48,000 tons untouched

Da Silva’s Gold Papers: Sibanye's plan to flood the richest reef on earth - at best gold price ever

Dr Duarte da Silva names SA mining's builders, leavers and breakers. Brilliant.

Duarte da Silva's Gold Papers: Who held the gold? - the numbers