Pan African’s Soweto DFS: Can a 29.55% IRR Survive Scrutiny?

A 29.55% internal rate of return is the kind of headline number that ends most feasibility study discussions before they start. For Pan African Resources’ Soweto Tailings Retreatment Project, it is where the discussion should begin, because that return rests on a gold price of US$3,550/oz and a capital bill that was reduced by R718 million during the study’s own value engineering.

The timing makes the question urgent. The definitive feasibility study landed on 11 September 2026, a final investment decision is targeted for December 2026, and the principal environmental authorisations the project needs are still pending, now expected in FY27 rather than mid-2026 as earlier guidance suggested.

For anyone weighing Pan African Resources right now, this is a real decision point, not a hypothetical.

This analysis works through the project in sequence: its capital structure, its cost and return profile, its permitting status, and its place in the group’s production build. The aim is to give you what you need to judge whether the Soweto economics hold up under scrutiny, rather than taking the headline figures on trust.

What the Soweto DFS actually promises: production, capital, and the cost of a bigger build

The Soweto Tailings Retreatment (STR) Project is a 600,000 tonnes per month circuit that reprocesses historic mine dumps west of Johannesburg, on tailings storage facilities Pan African acquired through the Mintails transaction. Its defining feature is that it does not stand alone.

The circuit plugs into the operational Mogale Tailings Retreatment (MTR) plant, borrowing its downstream processing infrastructure rather than building duplicates. That shared backbone is what keeps upfront capital lower than a standalone plant would demand, and it covers:

• Elution (stripping gold from carbon)
• Carbon regeneration
• Electrowinning (recovering gold onto electrodes)
• Smelting

The DFS also includes a dedicated new tailings storage facility (TSF) built to the Global Industry Standard on Tailings Management (GISTM), the international benchmark for dam safety introduced after a series of high-profile failures. This is the detail that matters most for interpreting the numbers.

The Global Industry Standard on Tailings Management sets out the engineering, governance, and emergency preparedness requirements that a compliant new tailings storage facility must meet, which explains why adding a dedicated GISTM-compliant TSF lifts the capital estimate so substantially relative to earlier, leaner project configurations.

That new TSF is the primary reason the capital estimate now sits at R3.68 billion, well above the roughly US$160 million configuration described in the earlier prefeasibility work. Value engineering trimmed R718 million from an initial higher figure while holding throughput and output targets steady, but the fuller scope still lifts the capital base materially.

The practical consequence: any cost or return figure quoted from the earlier PFS relates to a leaner, differently scoped project. Treat those numbers as non-comparable, not as a like-for-like baseline.

The gap between a prefeasibility and a definitive study is one of the most consequential distinctions in mining feasibility studies, because scope changes, not cost inflation, routinely produce the largest capital step-ups between study phases.

On production, the reserve underpins a long, steady life. The Soweto Cluster holds a mineral reserve of 108 Mt at 0.28 g/t, containing 0.98 Moz of gold, feeding output of 35,000-40,000 oz/year across 15 years for a life-of-mine total of 561,000 oz.

The strategic payoff is not the standalone figure but the combined one.

Once STR runs alongside MTR, peak output from the Mogale complex is expected to reach approximately 100,000 oz/year, effectively doubling the complex into a second West Rand production pillar.

IRR, NPV, and payback: reading the headline metrics with the right context

At the assumed gold price, the returns are genuinely attractive. The DFS puts the post-tax net present value (NPV, the value of future cash flows in today’s money) at R1.85 billion, roughly US$109 million, with a real ungeared IRR of 29.55% and a payback of approximately three years post-commissioning, all calculated at US$3,550/oz.

That gold price assumption is the pressure point. It sits near the top of the historical range, so the durability of these returns depends on how the economics behave if gold retreats from multi-decade highs.

Anchor to this number, not the old one The DFS all-in sustaining cost (AISC) is US$1,750-1,800/oz, excluding potential renewable energy savings. Earlier PFS-era figures of US$1,000-1,200/oz describe a different, lower-capex design and should not be read as the current cost base.

The cost step-up is not inflation within one design; it is a wholly different project scope, driven largely by the new dedicated TSF. Anchoring to the earlier figure would flatter the margin picture.

On sensitivity, the prior PFS work offers directional guidance despite its smaller capital base. That earlier configuration modelled an NPV near US$129.7 million and an IRR of 29.4% at US$2,800/oz, rising to an NPV of about US$235.4 million and a 40.2% IRR at US$3,500/oz. The read for you: returns move sharply with the gold price, but the project stayed positive well below the DFS assumption, which suggests the economic logic does not collapse the moment gold softens.

How the Soweto numbers compare with Mogale

Pan African’s own Mogale DFS is the most relevant internal benchmark, because both projects use the same integrated tailings model.

The temptation is to read Soweto’s higher IRR as the stronger project. Be careful. Mogale’s 20.1% was struck at US$1,750/oz and Soweto’s 29.55% at US$3,550/oz, so much of that gap is the price deck, not superior underlying economics. Mogale actually carries the lower AISC. What Soweto adds is scale, a longer life, and the complex-wide output uplift.

The permitting gap and what it means for the December FID

The December FID is best understood as a conditional milestone, not a firm date. Pan African has been explicit that the decision hangs on three things:

• Board approval
• Project financing secured
• Receipt of all required statutory authorisations

That third condition is where the schedule risk concentrates. In the 27 November 2025 update, the company expected the environmental impact assessment (EIA) and water-use licence approvals by June 2026. The September 2026 DFS moved the expectation for principal authorisations into FY27, a slippage of roughly six to twelve months against the earlier guidance.

Some pieces are already in place. Pipeline servitude approvals have been obtained, and the Integrated Water Use Licence Application (IWULA) has been submitted and is in progress.

But the full regulatory scope is broader than a single licence, and the project is seeking a 20-year authorisation period across:

The five separate approval streams Soweto must navigate reflect overlapping authorisation requirements that are common across major mining jurisdictions, where environmental, water, waste, radiation, and air-quality licences are administered by different agencies on different timelines, creating compounding schedule risk that project teams cannot fully control.

• Environmental authorisation
• Integrated water use licence (IWULA)
• Waste management approvals
• Radiation and nuclear registration
• Air-emission licences

For you as an investor tracking the name, permitting is not a technical footnote. It is the single variable most likely to determine whether the December FID holds or slides into 2027, and by extension when the first production ounce arrives.

Where Soweto fits in Pan African’s production build and what investors are buying into

Step back from the project and the picture changes. Soweto’s 35,000-40,000 oz/year is meaningful, but it is one contribution within a group trajectory that has been climbing steadily and is guided to keep climbing.

To size Soweto’s contribution, you need the denominator it sits against. The existing asset base already carries the bulk of production:

Pan African’s production trajectory over the past two reporting periods provides the operational context that makes the group-level FY31 projections credible: a 51% surge to 128,000 oz in the first half of FY26 shows the existing asset base performing ahead of pace even before the growth projects contribute.

• Elikhulu Tailings Retreatment Plant: approximately 50,000-56,500 oz/year
• Mogale (MTR): approximately 50,000 oz/year incremental
• Barberton Tailings Retreatment Plant (BTRP): approximately 13,000 oz/year
• Barberton Mines (underground)
• Evander Mines (underground)

Edison Investment Research projects group output reaching roughly 334,000-335,000 oz/year by FY31, but only once Soweto, Mogale, and the underground expansions are all delivering. That is the crux of the investment: you are not buying Soweto in isolation. You are buying a company whose stated growth depends on several concurrent projects executing broadly on plan.

The Emmerson acquisition adds another dimension to Pan African’s concurrent execution dependencies, layering an integration task onto the Soweto FID, the Mogale ramp-up, and the underground expansion programmes that all sit on the FY27-FY29 critical path.

Rising group AISC and what it means for Soweto’s margin buffer

The cost trajectory complicates the growth story. Group AISC guidance has been revised upward, with FY26 cited at up to US$1,575/oz in operational update coverage, and Alliance News reporting FY27 group AISC guidance of US$2,075-2,175/oz.

That is a steep climb in a single year, reflecting pressure across power, reagents, and logistics.

Set Soweto’s project-level AISC of US$1,750-1,800/oz against that rising backdrop and the margin buffer looks conditional. It assumes two things hold simultaneously: that gold stays near the US$3,550/oz DFS deck, and that cost inflation does not accelerate further at the project level. If either slips, the comfortable-looking spread narrows quickly.

Making the call on Soweto: what a disciplined investor weighs before December

Pull the four threads together and the trade-off is clear. Soweto offers a compelling IRR at current gold prices and infrastructure leverage that keeps capital intensity down. Against that sit a meaningful capital and cost step-up from earlier configurations, a permitting timeline that has already slipped, and a portfolio in which this is one of several concurrent execution dependencies.

Rather than a verdict, here are the three variables worth watching, in order of analytical weight:

1. Principal environmental authorisations. These are expected in FY27 and are a stated FID condition. Movement here, in either direction, is the clearest signal on whether December holds.
2. Project financing. No committed facility detail has been disclosed, and financing is the second FID condition. An update materially de-risks the decision.
3. The gold price versus the US$3,550/oz DFS assumption. The 29.55% IRR and roughly three-year payback are anchored to that deck. Sustained weakness compresses both.

A concrete anchor for your timeline Payback runs at approximately three years post-commissioning, and with commissioning around mid-to-late 2029, the return clock does not really start until the end of the decade.

None of this negates the strategic case. The infrastructure leverage is real, the environmental remediation of longstanding West Rand liabilities is genuine, and the lift in Mogale complex output to roughly 100,000 oz/year is a material scale gain. The task is holding those advantages and the risks in the same frame.

The reader who has followed this far should be able to state precisely what must be true for Soweto to deliver its DFS case, and to separate the conditions within Pan African’s control from those, chiefly the gold price, that are not.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.

Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors, and forward-looking statements are speculative and subject to change based on market developments and company performance.