← back to picks
Founded 1888 (London — modern PAF since 2000 reverse takeover) Reporting USD · FY-end 30 Jun Credit Aaa/AAA (net-cash) Valuation 2026-05-24 FCFF · scenarios on long-term gold price FCFF · Dark v3

Cycle-peak earnings · structurally undervalued at mid-cycle gold · -43.9% margin of safety

Market above mid-cycle DCF

Market $1.89/sh trades above mid-cycle DCF intrinsic $1.31/sh: at scenarios.yaml's bull case ($1.83/sh, current gold sustained) the stock is fairly priced; at the bear ($0.60/sh, full Edison $1,866/oz reversion) it is overpriced 3×.

bear $0.60 base $1.31 bull $1.83 MARKET $1.89 $0.40 $2.20
Sector Precious Metals Geo SA 80 · AUS 20 Listing AIM + JSE · USD reporter Credit Aaa/AAA (net-cash by Feb 2026) ROIC TTM 29% · 5y avg 16-18% Governance BEE-complex · 5% haircut
Intrinsic / share
$1.31
post 5% gov haircut
Market / share
$1.89
149p × GBP/USD 1.27
Margin of safety
-43.9%
vs intrinsic
Enterprise value
$2.88B
50.5% terminal
Cost of equity / debt
8.02% / 3.36%
β 0.52 · CRP 3.12%
Terminal ROIC / g
10.00% / 4.20%
spread ~212bp

What it sells, where it sells

Operating segments

283koz FY26 guide
Evander + Elikhulu (SA)Underground gold + tailings · mid-cost AISC ~32%
Barberton (SA)High-grade underground · world's oldest operating gold mine ~28%
MTR tailings (SA)$2.8M acquisition cost; ~$1,700/oz AISC; payback <2y ~20%
Tennant (AUS)First gold Q1 FY26; ramping to 50koz/y; Warrego Stage 2 by FY29 ~20%

Low-cost surface ounces (MTR tailings + Tennant) climbed 41% → 52% → 58% of group production FY24 → FY26 — the operating-margin debate is really a production-mix debate, not a gold-price debate.

Country mix (revenue-weighted CRP input)

🇿🇦South Africa (assets)80%
🇦🇺Australia — NT (Tennant)20%
🇬🇧UK (domicile only)0%

Weights are asset-located, not revenue-located — gold is a globally USD-priced commodity sold to LBMA refiners, so revenue location is meaningless. Weighted CRP ≈ 80% × 2.56% (SA) + 20% × 0% (AUS) ≈ 2.05% — load shedding and ZAR vol drive the spread, not customer demand.

Background — five things to know before the case

Context the share price doesn't carry on its face. Skim this once and the bull / bear bullets below stop reading as inside baseball.

Where we diverge from the sector

Each row asks: if we'd used the sector median for this one assumption instead of our override, how much would the share price change? Negative = our override is more conservative (worth less); positive = more aggressive (worth more). Skim down the rightmost column to see which bets are doing the work.

Assumption
Our input
Sector median
Value impact
01 Revenue growth (10y CAGR) Y1-Y5 then linear decay to terminal
~4.8% blended project-pipeline volume (Tennant + Warrego + Soweto), NOT gold price
~3.0% Precious Metals, global industry standard
+$0.12
02 Operating margin (Y7 target) Glide path from TTM 43.9% peak
28.0% mid-cycle anchor — DOWNWARD recalibration from peak
24.9% Global Precious Metals industry standard (itself cycle-contaminated)
+$0.08
03 Sales-to-capital (Y1-5) Revenue per $ of new capital
1.50 above industry but below TTM peak 1.03; Warrego + Soweto capex being capitalised
0.92 Precious Metals, global industry standard
+$0.10
04 Terminal growth Steady-state past Y10
4.20% = USD risk-free (Damodaran ceiling); captures gold-price inflation, not unit growth
4.20% industry standard ceiling (= rf)
±$0.00
05 Cost of capital (WACC) β override changes the discount rate
7.87% β = 0.52 (regression — gold's negative correlation in risk-off regimes)
9.50% β = 0.85 bottom-up industry standard
+$0.18
Net effect of overrides
Overrides net +$0.48/sh vs all-defaults — the β override does the heaviest lifting, reinforced by the higher S2C and margin overrides. Even with all overrides ON, intrinsic ($1.31) sits below market ($1.89), so the BUY case requires gold to stay above mid-cycle.
+$0.48
Our override Sector median Override adds value Override subtracts value

The story & the five claims

The 10-year story this DCF is built on, plus the five anchor claims that translate the story into the five DCF inputs.

The 10-year story

Pan African is a mid-cap South African gold miner with a ramping Australian foothold. Roughly 80% of FY26 ounces come from Barberton, Evander/Elikhulu and the MTR tailings operation, the remaining 20% from Tennant in the Northern Territory. The trailing twelve months sit at an unambiguous cycle peak — realised gold at $3,812/oz produced a 43.9% operating margin and a 29% ROIC, both well above any defensible long-run number. The base case isn't that the gold price collapses. It's that revenue grows 5% a year for five years from disclosed pipeline (Tennant ramp, Warrego Stage 2 by FY29, Soweto Cluster DFS in June 2026), then decays to the USD risk-free rate by Year 10, while the operating margin re-anchors at 28% mid-cycle by Year 7. Revenue compounds from $838M today to about $1.34B by FY35; terminal growth pins at 4.2%. The valuation lives or dies on the long-term gold price assumption — read scenarios.yaml alongside this base case.

Two debates worth pressure-testing

Is 28% terminal op margin defensible at mid-cycle gold?
Our view: 28% sits at the midpoint of Global Precious Metals (24.9%) and PAF's structural low-cost-tailings advantage (~88% of production at $1,700/oz AISC). A full Edison-style reversion to $1,866/oz real would drop margin to 15-20% and cut intrinsic to $0.60/sh (bear case). Wrong direction = ~50% equity hit.
Why use regression β 0.52, not bottom-up industry β 0.85?
Our view: Two-of-three regressions (Yahoo 5Y, SimplyWallSt) converge at 0.52; gold's negative correlation with broad equities in risk-off regimes is exactly what regressions capture and what bottom-up misses. Reverting to β 0.85 lifts WACC ~160bp and cuts intrinsic by ~$0.18/sh — large enough that the BUY case partially depends on this call.
CLAIM 01 Revenue grows from volume, not gold price. growth_high 5.0% · terminal 4.2% Tennant ramping to 50koz/y, Warrego adds 50koz + 10-15kt Cu by FY29, Soweto adds 35koz/y post-DFS. Y10 revenue ~$1.34B.
CLAIM 02 Mid-cycle 28% margin, not TTM 43.9% peak. target_op_margin 28.0% · convergence Y7 Midpoint of Global Precious Metals 24.9% and PAF's low-cost tailings advantage. Multi-year glide path — gold cycles don't mean-revert in one year.
CLAIM 03 Sales-to-capital 1.5 → 1.2 as capex normalises. s2c Y1-5 1.50 · Y6-10 1.20 Above industry median 0.92 reflects Warrego + Soweto capex cycle. Implied terminal ROIC ~24.5%, well above any reasonable WACC.
CLAIM 04 Net-cash balance sheet, zero failure probability. failure_prob 0.0 · terminal_g 4.2% Net-cash by end-Feb 2026. Even Edison's $1,866/oz bear generates positive cash from the lower-cost 88% of production.
CLAIM 05 ESO overhang priced; SBC inflated AISC ~$80/oz. options ON · 35M outstanding · 0.85 strike H1 FY26 AISC bridge attributed ~$80/oz to SBC escalation from the 140% share-price rise. Black-Scholes value subtracted from equity pre-per-share.
Where we diverge from sell-side
  • Mid-cycle margin re-anchor (28%) replaces TTM peak (43.9%). Sell-side models commonly extrapolate H1 FY26 economics 10 years out — we explicitly downward-recalibrate so Y7+ math sits inside historical PAF mid-cycle range.
  • Regression β 0.52, not bottom-up β 0.85. Two-of-three market regressions converge at 0.52 against an industry bottom-up of 0.85 — gold equity's negative correlation in risk-off shows up in price data, not in cross-section averages.
  • Treasury-share denominator honesty. 306.4M shares held by PAR Gold BEE vehicle stripped from the per-share base. Gross-share-count models overstate float by ~13%.
  • Volume-only growth driver — gold price is a scenario lever. 5% revenue CAGR sourced from disclosed projects (Tennant, Warrego, Soweto), not from price appreciation. Avoids compounding the cycle-peak distortion into the explicit forecast.
  • 5% governance haircut despite best-in-class M&A. Could argue 0.00 on M&A discipline alone; BEE-structure complexity and insider-selling pattern (CEO collar March 2026, Chairman 1M shares at 157p April 2026) deserve a modest reservation.
Two-sided case — bear anchors
  • Full Edison reversion to $1,866/oz real long-term gold. Operating margin compresses to 15-20%; failure probability flips from 0 to 5%; intrinsic drops to $0.60/sh (bear scenario) — a 68% per-share haircut from base.
  • ZAR strengthens further on commodity cycle. H1 FY26 saw 6.1% ZAR appreciation cost ~$115/oz of AISC. Another 10% ZAR strength shaves ~5pp from operating margin at mid-cycle gold.
  • Warrego or Soweto Cluster slips or runs over budget. FY27 capex $267M is fully funded today but a 30% cost overrun on either project erases two years of dividend capacity and undermines the 5% Y1-5 growth assumption.
  • Insider selling proves to be distribution-into-strength, not lifecycle profit-taking. No open-market buys during the 25% pullback would have been the reassuring counter-signal. CEO collar in March 2026 with attached loan is the quiet hedge at peak.
  • South African operational shock — load shedding or labour. Eskom power cuts directly cost ounces; an extended NUM/AMCU strike at Barberton or Evander could remove 5-10% of group production for two-plus quarters.

Risks to thesis (tail, not bear case)

Sub-$1,200/oz gold catastropheHigh

Below Edison's $1,866 anchor lies a tail scenario where even low-cost ounces struggle. Not contemplated in scenarios.yaml because it would render the whole sector uninvestable, not just PAF.

BEE structure regulatory shiftMed

SA Mining Charter changes or B-BBEE point-system tightening could force restructuring of PAR Gold / Concrete Rose / K2015 vehicles. Pure governance risk, not operational.

Tennant ramp underperforms guidanceMed

First gold poured Q1 FY26 but full 50koz/y nameplate not yet hit. Greenfield Australian operations are a different operating culture than the SA mature underground portfolio.

Tailings dam failure (TSF environmental)Med

~58% of FY26 production is tailings-retreatment. A Brumadinho-style failure at MTR or Elikhulu would carry catastrophic cleanup + reputational liability beyond insurance coverage.

AIM-to-LSE-Main-Market re-listing frictionLow

Persistent rumour PAF could uplift to Main Market. Forced index inclusion/exclusion mechanics around any re-listing event could create technical selling pressure unrelated to fundamentals.

Activist or short campaign on governanceLow

Currently zero short interest, no activist campaigns. The BEE complexity + insider-selling pattern is the kind of file that could attract an opportunistic short report in a sector pullback.

10-year forecast

Revenue $880M → $1.34B over 10y (5% Y1-5 from project pipeline, decaying to USD risk-free 4.2% by Y10); operating margin glides from TTM peak 43.9% to mid-cycle 28.00% by Year 7.

Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 0 1.3B rev (USD) 0% 45% op margin revenue FCFF op margin
⚠ terminal_growth (0.0420) >= risk_free_rate (0.0420); Damodaran's stable-growth ceiling is the risk-free rate.
Cost of capital build
Risk-free rate 4.20%
Mature-market ERP 4.23%
Levered β 0.52
Weighted CRP 3.12%
Cost of equity 8.02%
Pre-tax cost of debt (synth Aaa/AAA) 3.36%
D / V ~15%
WACC 7.87%
Full year-by-year DCF
Year Revenue Op mgn EBIT EBIT(1−t) Reinvest FCFF PV
1 $880M 41.61% $366M $254M $28M $226M $209M
2 $924M 39.34% $363M $252M $29M $222M $191M
3 $970M 37.07% $360M $249M $31M $218M $174M
4 $1.02B 34.80% $354M $246M $32M $213M $158M
5 $1.07B 32.54% $348M $241M $34M $207M $142M
6 $1.12B 30.27% $340M $238M $45M $193M $123M
7 $1.18B 28.00% $330M $234M $47M $187M $110M
8 $1.23B 28.00% $346M $247M $46M $201M $109M
9 $1.29B 28.00% $361M $261M $46M $215M $107M
10 $1.34B 28.00% $376M $275M $45M $230M $105M
Methodology & flags

Damodaran FCFF DCF, 10y explicit + perpetuity, base-year TTM to Dec 2025. R&D and lease capitalisation OFF (gold miner: exploration capitalised under IFRS 6; lease liabilities already in book debt). ESO overhang priced via Black-Scholes (35 M options outstanding, ~1.7% of net float). Country risk weighted SA 80% / AUS 20% asset-based. Synthetic credit Aaa/AAA reflects net-cash balance. Engine v1.0.0 · result: /Users/valentin/Documents/notes_jiliac_labs/Finance/Damodaran/valuations/paf/output/2026-05-25-result.json

  • R&D cap: OFF · Lease cap: OFF · Failure: OFF · ESO: ON
  • Governance haircut: 5% ($1.38 > $1.31)
  • Sensitivity tornado: not run (scenarios.yaml supersedes — bear/base/bull on long-term gold price)