Skip to content
Aarit Shah
Markets, Explained

Day 2

Why gold fell during a war

The safe haven asset dropped exactly when the textbook said it should rise. The reason is liquidity, and it explains more crashes than fear does.

· 5 min read

Gold is supposed to go up when the world gets frightening. Often it does. And then there are the days when conflict escalates, the headlines are as bad as they get, and gold falls anyway. People treat this as a paradox. It is not. It is a rule about how selling works.

In a genuine panic, correlations converge on one. Assets stop trading on their own merits and start trading on a single question: what can be sold right now without moving the price too much. Gold is deeply liquid. That is normally its virtue, and in a margin call it becomes its vulnerability. When a fund needs cash today, it does not sell the position it most wants to exit, it sells the position it can exit. Quality gets liquidated precisely because quality is sellable.

Layer on the second mechanism: gold pays no yield. Its appeal is inversely related to what you give up by holding it. When a crisis pushes expectations toward higher rates, or drives money into the dollar as the reserve asset of last resort, the opportunity cost of holding a metal that produces no income rises. A stronger dollar alone can push the gold price down while nothing about the geopolitical situation has improved at all.

So three forces act at once, and they do not point the same way. Fear pulls gold up. Forced liquidation pulls it down. Currency and rate expectations pull it in whichever direction the dollar went. On any given day, the one that wins is usually the most mechanical one, not the most emotional one, because forced selling has a deadline and fear does not.

The general lesson is worth more than the gold example. The reason an asset usually moves and the reason it is moving today are frequently different, and the second one is often plumbing rather than narrative: who is being forced to do what, by when. Narrative explanations feel satisfying and are available instantly. Plumbing explanations arrive late and are usually correct.

When something behaves the opposite of how it is supposed to, the useful instinct is not to conclude the market is broken. It is to ask who was forced to sell, and what deadline they were selling into.

Related

Building something in this space?

I build AI tools and trading infrastructure, and write this series alongside it.

Work with me