Platinum crashed 42% from its record high after a surplus forecast spooked investors

Platinum touched $2,920 an ounce in January, a record that erased a 17-year-old high. Nine months later it’s trading near $1,700, and the reversal says more about how fast speculative money can leave a thin metals market than it does about any mine running out of ore.

On January 26, platinum futures hit $2,925 an ounce, smashing the old 2008 record of $2,308.80 and capping a rally that had investors calling it the next gold. By September 1 it had fallen to $1,796.27, a 38% drop, and it has kept sliding since. As of early October, platinum sits around $1,699 an ounce, down roughly 42% from the January peak and more than 20% for the year, according to market data cited by XTB.

The culprit isn’t a sudden glut of mine supply. It’s a forecasting reversal that pulled the rug out from under the trade that drove the rally in the first place.

For most of 2025 and early 2026, the platinum story was simple. The World Platinum Investment Council had projected a 297,000-ounce deficit for 2026, the fourth straight annual shortfall, and banks piled in on the back of it. Bank of America Securities raised its 2026 average price forecast to $2,450 an ounce from $1,825 in January. It kept pushing targets higher after that, eventually calling for platinum to hit $3,000 an ounce by the fourth quarter, according to reporting from Mining Weekly and Kitco.

Then the WPIC changed its mind. In its most recent quarterly update, the council flipped its 2026 call from a 297,000-ounce deficit to a 265,000-ounce surplus, the first projected annual surplus since 2022. Mine supply barely moved, holding close to 5.55 million ounces. What actually shifted was investment demand, which the WPIC now expects to fall 18% this year. Investment flows didn’t just slow, they reversed: platinum-backed funds pulled in 1.15 million ounces in 2025 and are on pace to see 83,000 ounces walk out the door in 2026.

Rates didn’t help. The Federal Reserve raised rates in September for the first time since 2023, and the 10-year Treasury yield climbed near 5.14%, its highest in roughly two decades. Platinum pays no yield, so every basis point higher on a risk-free Treasury makes holding bars and ETF shares a little less appealing. Platinum’s 2026 rally was built almost entirely on investment flows rather than new industrial buying: that was enough to tip the balance.

The jewelry story that was supposed to carry 2026 is cooling too

Part of the original bull case rested on China, where platinum jewelry had become a cheaper stand-in for gold once bullion prices ran too high for ordinary buyers. It worked, for a while. Chinese platinum jewelry fabrication rose 56% in 2025 to 589,000 ounces, the strongest year since 2022, while gold jewelry demand in the same market fell 25%, a trade-down pattern precious metals analysts had been watching closely.

That momentum has faded. Global platinum jewelry demand is now expected to contract 6% in 2026, driven mainly by softer Chinese buying after the first-half surge ran its course. The gold-alternative trade hasn’t disappeared: gem-set platinum jewelry is still outperforming gold jewelry by a wide margin. But it was never going to offset a swing of more than a million ounces in investment demand on its own.

None of this erases the structural case that got banks excited in the first place. Above-ground platinum stockpiles still cover only about four months of global demand. Mine supply is still sitting near a 12-year low with little new investment behind it, and the WPIC itself still expects demand to outstrip supply in most years through the end of the decade on a cumulative basis.

What changed is simpler, and less comforting for anyone who bought near the top. A one-year surplus forecast, built almost entirely on investors deciding to sell, was enough to erase nine-tenths of a historic rally in under eight months. Platinum’s scarcity story may still be real.

Its price, for now, is being set by who’s selling, not who’s digging.