Capital expenditure picked up in the final quarter (to
US$12.5m) as expected, to end FY26 at
US$31.8m, 1.8% down compared to FY25. The company still expects a meaningful reduction in spend
over the coming years, with
US$9.2m guidance for FY27, of which
US$6.4m relates to tailings and infrastructure projects.
During Q426, the cash balance increased by 6.2% from
US$63.3m to
US$67.2m, with strong operating cash flow and favourable working capital movements exceeding
the higher capital expenditure, tax obligations and
US$7m paid as an interim dividend.
4E ounces per 1,000 issued shares more than doubled since FY13
Sylvania has grown 4E production per 1,000 shares in issue from 148oz in FY13 to 369oz
in FY26, an increase of 149% over a period in which plant feed rose only 31%. The
superior growth in production per share is thanks to the return on a sustained internal
investment programme, as well as capital management. Plant recovery improved from
c 40% in FY13 to c 57%, driven principally by the Project Echo secondary milling and
flotation modules, the share of plant feed routed to the PGM circuit rose from c 46%
to c 56%, and the continuous share buyback programme has reduced shares in issue by
c 13% since FY15.
| Exhibit 2: 4E PGM production expressed per 1,000 issued Sylvania shares |
 |
| Source: Edison Investment Research, Sylvania Platinum data |
Production per share has therefore compounded at more than twice the rate of production
itself. We view this record as central to the investment case, as it demonstrates
management's ability to grow the operating base independently of the PGM price cycle.
When the PGM cycle is added into the mix, it amplifies Sylvania’s significant delivery
as EPS increased 20-fold over the period from FY13 to FY26 (a compound annual growth
rate of 26%). There was a structural bear market in PGMs until FY16 (the average PGM
4E basket increased 2.5x from FY13 to FY26). We will explore these production, price
and capital management dynamics in greater depth in an outlook note in later in the
year.
Conservative production forecasts driven by medium-term SDO outlook
Sylvania’s SDO production has grown strongly over recent years, despite growing pressure
on plant feed, which peaked at 2.56m tons in FY25. After a dip in PGM feed tons in
FY25 (all SDO-related), which was affected by significant investment production improvements,
combined PGM feed tons rose by an impressive 11.6% in FY26, with c 4.5% largely due
to improvements in feed-head grade (resulting in a healthy rise in PGM feed relative
to overall feed) but also supported by the maiden contribution from the JV (c 93,000t).
| Exhibit 3: JV pick-up to buffer against forecast SDO production decline |
 |
| Source: Edison Investment Research |
We forecast SDO plant feed to be largely maintained at c 2.5m tons during FY27 and
FY28, before it starts tapering off from FY29, on the back of feed declines for the
Lannex, Millsell and Doornbosch operations, reducing to c 1.5m tons from FY32, when
we forecast it to stabilise for the remainder of our explicit forecast period to FY40.
We forecast combined plant feed to rise to a peak of 3.0m tons in FY28 on the back
of the JV PGM ramp-up to full production.
We forecast combined PGM plant feed to hold at c 1.50m tons across FY27 to FY29, in
line with management guidance that some SDO operations ran slightly above capacity
in FY26, and we have trimmed our SDO PGM feed-grade assumption to 3.65g/t (from the
3.71g/t realised in FY26) to reflect the partial depletion of higher-grade third-party
feed. On this basis, 4E PGM production remains broadly in line with the FY26 record,
at c 94,100oz in FY27 (including the work-in-progress related to FY26) and c 91,800oz
in FY28, recovering modestly to c 94,000oz in FY29 as the feed mix normalises, before
the SDO run-off takes hold from FY30. Forecast production stabilises at c 65,000oz
from FY32 until FY36, after which we allow for production from the JV to start moderating
(in line with our 10-year life-of-mine assumption). Our 2E PGM and 6E PGM production
forecasts follow a similar trajectory.
We have injected more conservatism into our production forecasts than in our previous
modelling, to reflect the uncertainty of long-term production due to Sylvania’s dependence
on feed from host mines where mineral resource estimates are not available as well
as the current terms of the JV agreement being limited to 10 years. We have also taken
a more conservative approach around PGM plant-feed grade and recovery, limiting further
improvements from current levels. We believe that uncertainty around long-term production
and efficiency may play a role in how investors perceive Sylvania, and we consider
our lower forecasts as a more conservative base case against which attractive upside
scenarios can be tested.
In our upcoming outlook note, we will explore alternative scenarios including a JV
term extension, life-of-mine extensions for dump operations such as Tweefontein and
Millsell, the continuation and potential growth of current arisings from host mines
(including Mooinooi and Lesedi) and continued extraction efficiencies, especially
as it relates to dump operations.
PGM downgrade on Middle East volatility
In our February 2026 report, we increased our PGM price forecasts on the back of the strong rally since late-December
2025. While the PGM rally peaked in February 2025, prior to the outbreak of the Iran
war, elevated platinum and rhodium prices were sustained into mid-May 2026. Since
then, platinum (down 20%) and rhodium (down 15%) have been under pressure, resulting
in an update in our PGM forecasts. Ruthenium and iridium have remained strong and
have been less affected by our forecast downgrades.
| Exhibit 4: Edison updated PGM price forecasts (average June year-end prices) |
 |
| Source: Edison Investment Research, Austin Lawrence Gidon, LSEG Data & Analytics |
Our FY27 forecasts are cut materially across the 4E metals, with rhodium down from
US$16,259/oz to
US$8,500/oz and platinum from
US$2,968/oz to
US$1,700/oz, as the war-driven spike has partly retraced, and FY27 is already c two months priced
at spot. From FY28 we forecast a gradual recovery, with rhodium reaching
US$11,623/oz by FY30, as autocatalyst demand proves stickier than the electric vehicle (EV) narrative
implies and AI-related demand builds across the PGM complex. By contrast, our ruthenium
and iridium forecasts are upgraded, reflecting their structural deficits and demand
bases that are decoupled from the autocatalyst cycle.
Our long-term prices apply from FY31 and hold the FY30 levels broadly flat in constant-currency
terms, at
US$12,045/oz for rhodium,
US$2,000/oz for ruthenium and
US$7,198/oz for iridium. These long-term assumptions matter more than any near-term forecast,
as they govern 10 of the 14 years of our explicit forecast period and account for
c 60% of the valuation sensitivity to each metal. We believe they are defensible on
supply fundamentals. South African supply discipline continues to constrain rhodium,
while ruthenium and iridium have no primary production, producer stocks have been
depleted over five years and demand from AI data storage, semiconductors and proton
exchange membrane (PEM) electrolysis is growing structurally. We test the sensitivity
of our valuation to these long-term prices in Exhibit 7, and will publish a comprehensive
review of the rhodium, ruthenium and iridium markets in our forthcoming outlook note.
Supply and demand dynamics for PGMs
When the Iran war ends, we think that PGM prices across the board would strengthen
as we believe AI demand will become dominant in all the PGMs over the next five years.
We believe our forecasts, shown above, are therefore conservative. Following the outbreak
of the Iran war, PGM prices fell as oil prices rose, the rationale being that internal
combustion engine (ICE) vehicles would be more expensive to run while EVs would be
much cheaper. As most of the palladium and rhodium produced is used in autocatalysts
to restrict emissions and comply with stricter regulations in Europe and China over
the next five years, palladium and rhodium prices fell initially.
Rhodium, ruthenium and iridium have insignificant above ground stocks while platinum
and palladium have some months of these stocks left. All PGMs are forecast to be used
in considerable amounts in AI-chip manufacture and data-storage systems where demand
is rampant. Platinum is likely to be used in data storage, hard disc drives, AI-crystal
production, AI servers (which run at high temperatures), manufacturing humanoid robot
fingers that are more sensitive and increased high-end glassmaking. In the rhodium
market, where stocks are depleted, demand is now being driven not only by autocatalysts,
as demand returns for catalysed cars, but also by the demand for high-end high-quality
glass for bases for AI chips. Palladium will be used extensively in multilayer-ceramic-capacitor
production for AI applications, but as there are still several months of above-ground
stocks available, its price is responding at a slower rate. Furthermore, even though
there has been a shift to pure EVs in Europe because of the Iran war, EV sales are
lagging hybrid-vehicle sales, and hybrid vehicles use ICEs, meaning that catalysed
car sales are actually increasing in Europe and China. In the US, ICEs dominate and
EV sales have plummeted. However, ICEs still dominate outside the US, with the result
that PGM demand for autocatalysts is set to increase, with an extra lift from AI and
other applications over the next few years.
Large exposure to rhodium and growing importance of ruthenium and iridium
Exhibit 5 shows the estimated contribution of each metal to Sylvania’s realised PGM
revenue, after payability. Rhodium contributed 33–37% of realised revenue in each
quarter of FY26 despite being only c 9% of 6E ounces, which makes it the single largest
earnings driver. Ruthenium and iridium together contributed 11–15%, and their share
is growing, with ruthenium's contribution roughly doubling since early FY25 as its
price re-rated. We present these contributions net of payability. Sylvania realises
c 74% of the gross 4E basket but only c 45% of the value of its ruthenium and iridium
content, so ounce or gross-basket shares materially overstate the by-products’ contribution.
| Exhibit 5: Estimated contribution to 6E revenue (FY25 and FY26, quarterly) |
 |
| Source: Edison Investment Research, Sylvania Platinum data |
Improved payability on the minor PGMs, which management has targeted through the centralised
filtration plant, is therefore a source of upside that requires no additional production.
Together with chrome, which contributed
US$9.8m of revenue in FY26 and which we forecast to grow to c 15% of revenue in FY27 and
c 17% in FY28 as the JV reaches full production, the revenue base is far more diversified
than it was two years ago, and we will return to this theme in our forthcoming outlook
note.
Forecasts cut on lower PGM forecasts
Our FY27 revenue forecast has been cut by 40% to
US$238.0m and FY28 by 37% to
US$266.8m, driven principally by the lower PGM price deck and reinforced by our adoption of
management's flat near-term production guidance. FY27 normalised EPS has been cut
by 57% to 28.0 US cents, dipping below the 30.5 US cents delivered in FY26, before
the recovering price deck lifts FY28 to 33.6 US cents.
We forecast capital expenditure of
US$13.8m in FY27, declining sharply from
US$32.9m in FY26 as the Thaba JV build is complete. Our forecast sits comfortably ahead of
the company's initial
US$9.2m guidance (of which
US$6.4m relates to tailings and infrastructure projects), providing headroom should that
guidance be refined upward at the FY26 results. Lower capital expenditure, the commencement
of JV-related loan repayments and forecast healthy cash generation support a strongly
growing cash balance, which we forecast to rise from
US$67.2m in FY26 to c
US$100m in FY27 and c
US$161m in FY28.
During H126, Sylvania recognised a
US$12.3m impairment loss on its Hacra exploration asset, following a decision not to invest
further in the project. This is excluded from normalised earnings and reduces our
exploration assets valuation, which is based on the company's book value.
Sylvania's dividend policy is to pay a minimum of 40% of adjusted free cash flow,
with one-third at the interim and two-thirds at the final, and our forecasts now apply
this policy directly. For FY26 we forecast a total ordinary dividend of 6p/share,
more than double the 2.75p/share of FY25, comprising the 2p interim already paid and
a forecast 4p final to be declared with the FY26 results. On our lower price deck,
the policy yields 6.9p/share in FY27 and 9.6p/share in FY28, with the shares yielding
c 6.6% on FY26. With the cash balance building well beyond operational requirements
from FY29, we see increasing potential for windfall distributions over and above the
ordinary dividend. The company also launched a
US$2.0m share buyback programme during Q326, continuing a repurchase record that has run
since FY15, and we believe further buybacks remain a live mechanism for distributing
surplus cash.
