Cash generation and future development

PAF has indicated that, at current rates of cash generation, it will extinguish net debt by the end of February this year. Beyond that, we forecast that the company will continue to generate cash from operations comfortably above US$100m per year (and potentially around US$600m per year depending on the gold price), which will comfortably allow it to develop the 1,700km² of prospective, near-term, shallow Australian acreage that it owns in the Northern Territory (see Exhibit 6).

Valuation: £2 hoves into view

Our core valuation of Pan African has risen by 65.7% to 55.65c per share (cf 33.58c previously), based on its six producing mines. However, this uses a gold price that appears ever more conservative (see Exhibit 2). It rises by a further 35.00–40.02c (25.61–29.28p) to 90.65–95.67c (66.34–70.01p) if other assets, such as Egoli and the Soweto Cluster, are included. However, it more than quadruples, to 226.27c (165.60p), at the prevailing price of gold of US$5,000/oz at the time of writing. Alternatively, if PAF’s historical average price-to-normalised headline earnings per share (HEPS) ratio of 8.1x for the period FY10–25 is applied to our FY26 and FY27 forecasts, it implies values of 132.67p and 127.83p, respectively, or 220.93p in FY27 if the price of gold remains at current levels until June 2027. In the meantime, it remains cheaper than its principal medium-sized international gold mining peers on at least 69% of commonly used valuation measures if Edison forecasts are used or 65% if consensus forecasts are used (Exhibit 12). In fact, this peer group implies a PAF share price of 226.41p in FY26 and one of 140.12p in FY27 (or 242.20p in FY27 if the current gold price prevails until June 2027), based on Edison’s normalised HEPS estimates. This is further validated by a valuation of 95.59p on a cash flow and terminal multiple-type analysis (on an ex-real growth assumption), rising to 216.52p per share assuming real cash flow growth of 4.1% per year (which is the average real return of the gold price from 1967 to 2025).

H126 production results and outlook

On 26 January, PAF announced an operational update for the six-month period from 1 July to 31 December. Its main features were as follows:

• A 51% increase in gold production, from 84,705oz in the prior year period to 128,296oz, with further increases anticipated in H226 in order to meet full-year guidance (which was maintained) of 275–292koz. This follows record production of 111,822oz in H225.
• AISC of US$1,825–1,875/oz in H126. This is above erstwhile FY26 guidance of US$1,525–1,575/oz, albeit with the caveat that the latter was calculated at an fx rate of ZAR18.50/ US$, whereas the prevailing rate in H126 was ZAR17.37/US$ and the rate currently is ZAR16.0266/ US$.
• A 67% reduction in net debt, from US$150.5m in June 2025 to US$49.9m in December 2025 (December 2024: US$228.5m).
• A proposed (maiden) interim dividend of ZAR0.12 per share.

In terms of production, our interpretation of PAF’s announcement relative to our prior expectations is provided in Exhibit 1 below.

Stand-out features of the six-month period under review were the sharp (87.3%) increase in output at Evander underground and Elikhulu’s third-best semi-annual performance ever, partially offset by performances tracking the bottom end of the guidance range at Barberton, MTR/Mogale and Nobles.

Evander, in particular, benefited from full access to its sub-vertical shaft’s full 700t/day hoisting capacity, which has now been available since April, at the same time as it is mining the high-grade Kinross Channel of the Kimberley Reef. As a consequence of the increase in face length and mining flexibility occasioned by the establishment of the high-grade 24 Level B-Line raise at the start of this year, it has thus been able to maintain output of 3,607oz per month over the entire six-month period (cf 3,850oz pm in May and June).

At the same time, Barberton’s high-grade areas of the 262 Platform at Fairview Mine, indicated by drill intersections of up to 80g/t Au, have been accessed and should begin to make a positive contribution to grades in H226, while output at MTR/Mogale was approximately 10% below management’s expectation owing to mine sequencing. Nevertheless, expansion to 1,000ktpm was successfully commissioned in December 2025, with increased capacity and improved recoveries expected to increase gold production in H226.

Concurrently, after hot commissioning in April and an inaugural gold pour in May, production at Tennant Mines’ Nobles operation in Australia was hampered by a delay in the commissioning of the filter presses associated with the dry stack landforms (required for tailings deposition). Nevertheless, production in H226 is anticipated to increase to c 30,000oz as higher-grade, open-pit ore replaces lower-grade feed from the Crown Pillar Stockpile, such that the average expected recovered grade almost doubles from 1.15g/t to 2.22g/t.

The average gold price in H126 was US$3,802/oz (source: Bloomberg), while the average fx rate was ZAR17.37/ US$. Consistent with our general policy, our gold price forecast for CY26 now assumes that the current spot price of US$5,000/oz will prevail for the remainder of the calendar year (which is materially higher than our prior assumption), before reverting to our long-term levels as follows:

A note on gold prices

The gold prices in Exhibit 2 are derived with respect to historical precedent. However, almost the only modern precedent to today’s market is that of 1970–81 when gold rose from its post-war currency peg of US$35/oz to a peak of US$850/oz in January 1980 before falling by over 60% in the following two years. The analysis above implicitly assumes a repeat of the same pattern, with 2026 being an analogue to 1980 and 2027 being an analogue to 1981. However, there are material differences between the two periods of time. The biggest is that, in 1980, the US was still the world’s largest creditor nation and what suddenly reversed gold’s fortunes was the policy adopted by the then-new Federal Reserve chairman, Paul Volcker, to ‘defend the value of the US dollar.’ That entailed sharply raising real interest rates from near zero to around 4% (among other things), causing a sharp recession in the US and most other western countries in the early 1980s where they remained for almost the next two decades. Now, however, the US is the world’s largest debtor nation and no one in either the US administration or the Federal Reserve is talking the language of dollar defence. In fact, quite the opposite: they are talking about allowing the dollar to find a level at which US exports can compete on world markets and stimulating the domestic economy with real interest rates as low as possible. Hence, all the forces that have pushed gold to its current peak are still pushing it in the same direction (ie upwards).

While it is tempting to look at graphs of the gold price and to attempt to call a ‘top’, investors should be cautious as many of the forces that drive it are often self-reinforcing, especially the fact that above ground stocks of gold of c 216,000 tonnes dwarf newly mined supply of c 3,700 tonnes per year. Hence, traditional supply and demand analysis often fails in the case of gold, where price discovery tends to occur among existing holders, rather than new buyers. This means that, while the price has appreciated a lot, in the absence of a fundamental shift in macroeconomic policy, there is no reason to suppose that it cannot continue to rise for many years to come. The following demonstrates the extent to which this is possible:

• The gold price required to cover the total US monetary base is US$20,273/oz. This is analogous to the classical gold standard, according to which the Federal Reserve was required to hold enough gold to redeem all of its liabilities (ie US dollars) that could be in circulation. Although President Nixon formally closed this dollar window in August 1971, in the era of a floating gold price, US gold reserves were nevertheless still able to cover the US total monetary base as recently as 1980.
• The gold price required to cover the US net international investment position is US$109,416/oz. While this number appears very large, it would theoretically allow the US to cover all of its accumulated deficits since c 1979 and, to a certain extent, reset its economy (albeit with a currency unbacked by gold).
• The gold price required to cover the US net international investment position and to cover its monetary base is US$129,690/oz.

While gold would need to increase 26 times to get from its level now to US$129,690/oz, it is perhaps worth noting that it has already gone up by 143 times to get from its level of US$35/oz in 1967 to its current price. The main impediment to the last two scenarios will be the reaction of other (creditor nation) central banks. The world’s largest three creditor nations are Germany, Japan and China. While Germany and Japan already have currencies that freely float against the US dollar, China does not. With the US facing the prospect of a material decline in the purchasing power of the dollar in continuing to maintain its currency peg, China will also subject its citizens to a similar decline at a time when this is perhaps not their expectation. Therefore, at some point it is likely that the People’s Bank of China will abandon this currency peg in order to preserve its citizens’ wealth, albeit at the cost of accepting a much weaker US dollar and the US becoming much more competitive in global export markets. Of course, the US dollar decline may not go all the way to the levels discussed above and it is possible for the US and China to reach some sort of accommodation. However, this analysis also shows that, in the absence of a change in policy direction from either nation, the bull market for gold may be very far from over.

Edison forecasts

In addition to changes to our immediate production and gold price assumptions, we have also revised our estimate of fx rates to reflect the recent (slightly unusual) strength of the rand against both the US dollar and sterling:

• from ZAR22.4986/£ to ZAR21.9007/£ (-2.7%);
• from ZAR16.7557/US$ to ZAR16.0266/US$ (-4.4%); and
• from US$1.3425/£ to US$1.3664/£ (+1.8%).

In tandem with PAF’s updated cost guidance, we have therefore revised our operational forecasts for PAF’s mines for both H126 and H226 to those shown in Exhibit 3 below.

Given these operational assumptions, total cash costs for the group for the full year average out at US$1,590/oz, which is consistent with an AISC of US$1,769/oz. This is above PAF’s guided range of US$1,525–1,575/oz but it is largely consistent in rand terms (Edison ZAR951,191/kg vs PAF ZAR921,930/kg).

Due to these changes, we have upgraded our FY26 financial estimates for the group to those shown in Exhibit 4 below.

A comparison between Edison and consensus forecasts for FY26 is provided in Exhibit 5, below:

Group production

With the exception of FY26 (short term), we have made no changes to our medium- and longer-term production forecasts, with the result that, on current plans, we continue to expect PAF’s output to peak at 343.2koz in FY31, as shown below:

PAF’s accelerated development of the Soweto Cluster was considered in our previous note. Apart from that, PAF has a number of other potential organic growth projects in prospect, including:

• Fast-tracking the wholly owned Warrego gold and copper project at Tennant Mines with the objective of budgeting first production in 2029/30. The subject of a recent feasibility study, the Warrego project is targeting output of c 100,000oz gold and 10,000–15,000t copper per year over more than 10 years at a capital cost of US$40–45m. Regional gold and copper deposits owned by third parties could supply additional supplementary feed and a feasibility study to this end is currently in progress.
• A further potential project arising from the Tennant Mines acquisition is the White Devil project, where a recent scoping study commissioned by joint venture partner Emmerson Resources was reported to have delivered ‘encouraging results’, confirming an updated mineral resource of 4.6Mt at 4.2g/t gold, for 611,000oz contained gold, of which 87% is in the indicated category and therefore eligible for upgrade into reserves.
• Once the Barberton Tailings Retreatment Project’s tailings resources are depleted, management plans to convert the plant to process hard rock feedstock from the Sheba Fault project (comprising the Western Cross and Royal Sheba orebodies), which has an estimated mine life of nine years, with both orebodies open at depth.

Updated (absolute) valuation

Based on the present value of the estimated potential dividend stream payable to shareholders over the life of its mining operations (applying a 10% discount rate to US dollar dividends), our absolute valuation of PAF (based on its existing six producing assets and Edison’s long-term gold price of US$1,866/oz in real terms) has risen by 65.7% to 55.65c (cf 33.58c previously).

A summary of the major components of the change to our valuation is provided in the graph below. Readers’ attention is drawn to the relatively outsized effect that changes in the near-term price of gold have on our valuation:

As noted previously, however, this valuation is conducted at our relatively conservative gold price assumption of US$1,866/oz (real) for FY30 and beyond. At the current gold price of US$5,000/oz, all other things being equal, our valuation more than quadruples to 226.27c (165.60p):

Even at our lower, long-term gold price, however, including its other growth projects and assets, our updated total absolute valuation of PAF as a whole rises to 90.65–95.67c (66.34–70.02p):

Historical relative and current peer group valuation

Historical relative valuation

Exhibit 11 below depicts PAF’s average share price in each of the financial years from FY10 to FY25 and compares this with HEPS in the same year. For FY26 and FY27, the predicted share price is shown, given our forecast normalised HEPS for those years. As the chart shows, PAF’s price to normalised HEPS ratios of 8.9x and 9.1x for FY26 and FY27, respectively, remain only slightly above the average of its recent historical range of 4.1–14.8x for FY10–25.

If PAF’s average year-one price to normalised HEPS ratio of 8.1x for FY10–25 is applied to our updated normalised earnings forecasts, it implies a share price for PAF of 132.67p in FY26 followed by one of 127.83p in FY27 (as shown in Exhibit 11). Stated alternatively, PAF’s current share price of 144p, at prevailing foreign exchange rates, appears to be discounting FY26 and/or FY27 normalised HEPS of 24.35c per share (cf our forecasts of 22.19c and 21.62c, respectively).

Relative peer group valuation

In the meantime, PAF appears to remain cheap relative to a peer group of medium-sized, international gold miners on 69% of comparable common valuation measures (57 out of 82 individual measures in the table below) if Edison forecasts are used or 65% (54 out of 82 of the same measures) if consensus forecasts are used.

Alternatively, applying PAF’s peers’ average year one P/E ratio of 16.4x to our normalised HEPS forecast of 22.19c per share for FY26 implies a share price for the company of 266.41p at prevailing fx rates. Applying its peers’ average year two P/E ratio of 8.9x to our normalised HEPS forecast of 21.62c per share for FY27 implies a share price of 140.12p. However, our FY27 forecast is based partly on the current price of gold (for H127) and partly on our long-term price of gold (see Exhibit 2 for H227). At the current price of gold for the full year, our normalised HEPS forecast for FY27 rises to 37.36c, whereupon the share price implied for PAF by its peers (above) rises to 242.20p.

Valuing blue-sky upside

PAF is a multi-asset company that has shown a willingness and ability to grow production both organically and by acquiring assets to maximise shareholder returns. As a result, rather than our customary method of discounting maximum potential dividends over the life of operations back to 1 July 2026, in the case of PAF, we can alternatively discount forecast cash flows back over five years to the start of FY26 and then apply an ex-growth terminal multiple to forecast cash flows in that year (FY31) based on the appropriate discount rate.

In this case, our estimate of PAF’s pre-financing terminal cash flow in FY31 is 5.78c (at a real gold price of US$1,866/oz in current money terms). Applying a (real) discount rate of 5.92%, calculated from a nominal expected equity return of 9% and increased long-term inflation expectations of 2.2908% as defined by the US 30-year break-even inflation rate (source: Bloomberg, 27 January), to this estimate of cash flows, our valuation of the company is 95.59p per share in FY26 (cf 75.65p/share previously) assuming zero long-term cash flow per share growth beyond FY30.

At this point (FY31), production is anticipated to be c 343.2koz. If PAF is able to maintain this level of cash flows per share via organic investment, its valuation will flatten out at 71.41p per share in real terms on an ex-growth basis. However, the gold price alone should afford PAF’s cash flows an additional 4.1% per year growth in real terms (the compound average annual real appreciation rate in its price from 1967 to 2025), in which case PAF’s terminal valuation more than doubles to 242.18p per share and its current valuation to 216.52p per share (cf 135.75p/share in FY25 previously).

Financials

PAF reported net debt of US$49.9m on its balance sheet as at end-December 2025 (vs US$150.5m at end-June 2025 and US$228.5m at end-December 2024, which was the recent peak) and indicated that, at current rates, it would be extinguished by the end of February this year. Beyond that, we forecast that PAF will continue to generate cash from operations comfortably above US$100m per year (and potentially around US$600m per year depending on the gold price etc).