Thesis

The GLTR 10-K (FY2025) surfaces a platinum demand catalyst: Chinese consumers are substituting platinum for gold jewelry as the gold/platinum ratio hits historic extremes (2.41:1 as of March 8, vs historical platinum premium over gold). Four independent platinum miners confirmed this shift on Q4 2025 earnings calls. The Guangzhou Futures Exchange launched platinum futures in November 2025, creating institutional infrastructure for Chinese platinum investment. Supply remains structurally constrained — South African mines declined from 81 to 53 shafts over the past decade, and all four miners forecast continued deficits through 2026.

The demand catalyst is real. The supply constraint is real. But the factor decomposition says this is a sector bet dressed as an idiosyncratic thesis.

The Factor Problem

Regressing PPLT against GLD, SLV, SPY, and UUP over the trailing year:

PPLT = 0.24×GLD + 0.59×SLV + noise + ε

R² = 0.61 → Idio = 39%

SLV dominates (β=0.59, t=10.7). Platinum trades like silver, not gold — both are industrial-precious hybrids. When you include silver as a factor, gold drops to marginal significance. Sixty-one percent of PPLT's return variance comes from precious metals beta where nobody has edge.

The 39% idiosyncratic component is where the thesis lives. Here's what's in it:

Every structural driver in the idio component is either consensus or public. The Chinese demand story was discussed on four Q4 2025 earnings calls (SBSW, ANGPY, IMPUY, SAPLF), in WPIC research reports, at London Platinum Week, and on Bloomberg. When information appears in every miner transcript, it's priced.

What the Corroboration Actually Shows

The corroboration across four miners was strong — but in the wrong direction for alpha seekers.

Sibanye-Stillwater (SBSW, Q4 2025): "Jewelry manufacturers switched into platinum as gold too expensive. Platinum into China up 7% year-on-year, 4.5 million ounces." GFEX volumes hit 6M oz/day in December — exchange imposed trading restrictions.

Anglo American Platinum (ANGPY, Q4 2025): "Platinum rose considerably, still trading substantial discount gold. Enabled platinum jewelry gain market share several key geographies." Forecast "sizable deficit" in 2026.

Impala Platinum (IMPUY, Q4 2025): "Gold fatigue perhaps opportunity precious metal, platinum, as massive discount, record discount relative gold." Physical stock moving from Europe/US to China.

South African Platinum (SAPLF, Q4 2025): "Seen interesting demand increase jewelry, comes especially when gold current gold price."

Four CEOs on four earnings calls, all legally accountable, all saying the same thing. That's confirmation the thesis is correct. It's also confirmation the thesis is consensus. Every PM analyst on the planet heard these calls.

The bonus find: Pandora (PNDRY, Q4 2025) announced PANDORA EVERSHINE — shifting 75% of its silver jewelry assortment to platinum-plated over 2027-2029. World's largest jeweler by volume creating structural platinum demand independent of Chinese substitution. Incremental, but 2027-2029 timeline means it's not yet in demand forecasts.

The Gold/Platinum Ratio

This is the single most interesting number in the analysis.

Gold/platinum at year-end 2025: 2.12:1 ($4,308 / $2,027). As of March 8, 2026: 2.41:1 ($5,159 / $2,142). The ratio is WIDENING despite the Chinese substitution thesis — gold's central bank bid (23.6% of 2024 demand, highest on record) is outpacing platinum's industrial recovery.

Historically platinum traded at a premium to gold. The ratio has been inverted since roughly 2015. At 2.41:1, this is the most extreme inversion in modern history. WPIC estimates 0.7-1.5 Moz of long-term gold-to-platinum switching opportunity driven by price sensitivity in Asian jewelry markets.

The ratio tells you two things simultaneously: (1) the substitution pressure is intensifying — more extreme ratio means more price-sensitive consumers switch, and (2) gold has a structural monetary bid that platinum lacks. The ratio could stay extreme or widen further if central bank buying and de-dollarization continue to favor gold.

Platinum cannot close this ratio on its own merit. It needs gold to stop rallying, or it needs a catalytic event that reprices platinum's monetary/store-of-value utility. The GFEX launch was a step in that direction — but December's 6M oz/day volumes have a speculative overlay that SBSW's analyst explicitly warned will correct post-settlement.

Supply Deficit — Real but Known

The supply side is genuinely constrained and not fixable quickly:

• SA mine shafts: 81 (2008) → 53 (2025). No reversal possible in under 5-7 years.
• WPIC: 692 koz deficit in 2025 (third consecutive year). Projects narrow 20 koz surplus in 2026, but contingent on ETF outflows and trade easing.
• All four miners confirmed: structural constraints prevent materially higher supply, new mines unlikely at current prices.
• Recycling up 10% YoY but still below pre-COVID levels.

The deficit is real. But it's been the thesis for three consecutive years. The market priced platinum +122% in 2025. The information is fully in the public domain.

The Bear Case

Hydrogen demand collapsed. $543M+ in industry write-offs (CF, CMI/Accelera, BE). WPIC's 2030 forecast of 875-900k oz platinum from hydrogen now looks overly optimistic. CCS winning over electrolysis (45Q tax credit vs 45V). This demand pillar is impaired.

GFEX is partly speculative. SBSW analyst: "Going [to] see East moving up little bit as get closer [to settlement] date. Settlement essentially, going [to have a] nice cleverly made platinum stockpile [in] China. Post, get price correction." Investment demand ≠ structural demand.

WPIC projects surplus in 2026. 20 koz, narrow, contingent — but the authoritative industry body is calling a surplus, not a deficit.

Recycling responds to price. +10% YoY. At $2,100+, secondary supply economics improve. Price-elastic supply response caps upside.

What This Means

PPLT is not an alpha position. It's a factor bet with a platinum tilt. Buying it outright adds PM beta plus a 39% idiosyncratic component where the strongest drivers are consensus, not informational advantage.

At 39% idio, outright long PPLT is mostly levered PM beta. The clean expression for a platinum-specific view is long PPLT / short GLD to isolate the idiosyncratic component. The GFEX post-settlement correction is the most probable near-term catalyst for a better entry.

Predictions

Evidence