Perseus Mining: The Quarter Where Less Gold Made More Money
Output fell, cash didn't. With $900m in the bank and zero debt, Perseus has quietly retired every risk narrative it spent 2026 defending — and made capital return its loudest keyword.
PRU.AX · Earnings Call · 2026-10-26
Perseus Mining: The Quarter Where Less Gold Made More Money
The number that went the wrong way — and didn't matter
Perseus produced 92,551 ounces in the September quarter, down on June. The company's framing was immediate and, on the evidence, defensible: “We produced 92,551 ounces of gold, which was lower than the previous quarter, exceeded our plan” — Craig Jones, Managing Director and CEO · 2026-10-26. Planned maintenance shutdowns hit every asset — Yaouré's mill, Edikan's grade sequence (0.67 g/t), Sissingué's oxide-heavy Antoinette feed. And yet gold sold was 99,661 ounces, a clear draw on inventory, at a realised USD 3,881 an ounce, generating USD 171m of notional cash flow. The distinction between produced and sold is doing real work here, and management was transparent about where it bites: the company's own all-in site cost of USD 2,035/oz sits above the World Gold Council all-in sustaining cost of USD 1,855/oz, largely because of that sold-versus-produced basis and the deliberate build of lower-grade stockpiles at Yaouré as accelerated mining gets ahead of the mill. Both numbers are true. Only one of them will be quoted. The balance sheet did the rest of the talking: USD 900m in cash and bullion, zero debt, USD 400m undrawn, and a USD 1.3bn liquidity position that management pointedly noted excludes “$338 million of liquid investments in relation to our investments in PDI Gold and in Aurum” — Lee-Anne de Bruin, Chief Financial Officer · 2026-10-26. Full firepower is nearer USD 1.6bn. With gold guidance struck at USD 4,000/oz against a USD 3,881/oz realised price, the assumption is not stretched.Capital return took the top of the keyword table
For two years the dominant Perseus keyword was jurisdictional — Côte d'Ivoire, the Ivorian government, the additional royalty paid "in good faith," the new mining code. This quarter, return capital is the single highest-momentum keyword in the company's set, and it is no longer an aspiration. An AUD 100m special distribution (~AUD 0.0754/share) is being funded from the Sudan asset sale, structured as a capital reduction of up to AUD 70m with the balance as a special dividend — a structure clearly engineered for tax efficiency after an analyst on the August call pressed exactly that point about unfranked dividends and the new capital gains regime. The ATO class ruling is the swing factor: “An application to the ATO for a class ruling has been submitted on what portion will be declared as a capital reduction versus a special dividend.” — Lee-Anne de Bruin, Chief Financial Officer · 2026-10-26 AGM 30 November, record 18 December, paid 29 December. The market buyback is the weaker sibling. Of an authorised AUD 350m, only ~AUD 6.6m had been executed by quarter end:That is a 2% execution rate on an authorisation with a twelve-month runway. Blackout mechanics explain part of it; a share price near highs explains more. Watch whether the buyback becomes a real instrument or a permanent option.By the end of September, we had bought back just over 1 million shares for about AUD 6.6 million. And this takes into account the fact that we've had 1 month of nontrading due to trading blockouts.
The themes that quietly left the room
The most informative part of this quarter is what is no longer being said. In April, Lee-Anne de Bruin was answering fuel questions with real alarm about Ghana: “we're seeing sort of similar to what you're seeing in the Australian market in Ghana, where prices are going up probably 50%.” — Lee-Anne de Bruin, Chief Financial Officer · 2026-04-22 Diesel — then ~10% of all-in site cost — has vanished from the vocabulary entirely. Royalties have flipped sign. The 2026 fiscal negotiation was a threat; this quarter lower royalties is a top-five keyword and a stated driver of Edikan's 11% all-in site cost decline to USD 1,738/oz, alongside an over-accrual reversal from the prior quarter. The same line item that was a risk is now a tailwind. Hedging has gone quiet too. In October 2025 the company was buying puts at "between $40 and $70 an ounce" to protect below USD 2,600. With USD 900m of cash and no debt, downside protection is no longer a talking point — nor, notably, is there any fresh discussion of locking in USD 3,881 gold. And Nyanzaga's supply chain, a genuine analyst concern in August, was closed out then with “So as we stand today, no major issues.” — Craig Jones, Managing Director and CEO · 2026-08-25 This quarter the project is simply 88% built with procurement complete. Resolved problems stop being keywords.Nyanzaga at 88%, and an accounting wrinkle worth modelling
“The project is 88% complete and remains on budget and schedule for first gold in January.” — Craig Jones, Managing Director and CEO · 2026-10-26 USD 407m of the USD 483m budget spent, plus USD 33m of capitalised pre-production, pre-strip at 4.6m BCM against a 7–8m BCM January target, tailings dams ahead of schedule. Execution risk is now commissioning risk, which is a different and smaller animal. But there is a modelling trap in the guidance:Group guidance of 420,000–480,000oz at USD 1,835–2,070/oz includes 55,000 Nyanzaga ounces but none of its operating cost. Second-half group AISC therefore gets a mechanical assist that is an accounting artefact, not an operational improvement. Meanwhile Edikan cutback spend — USD 22.5m in the quarter, USD 27.6m cumulative, ~USD 140m for the year — is running through the easy free-dig upper benches; Craig Jones was explicit that costs rise as the pits deepen and drill-and-blast begins.Nyanzaga isn't included in our cost guidance as its operating costs are capitalized until commercial production in the June quarter of 2027.