So the queue outside my local Bangkok gold shop may have looked shorter, but somebody was still emptying the shelves through the loading dock.
Takeaways
-
’s relationship with appears to be changing. During the first phase of the Iran conflict, higher crude hurt bullion through inflation, Fed and real-yield channels. Gold is now showing signs of absorbing those same headwinds.
-
Central-bank demand looks considerably stronger once unreported sovereign buying is included, suggesting the structural reserve-diversification bid never disappeared.
-
China remains critical. PBoC accumulation, Shanghai physical withdrawals and Hong Kong flows are occurring alongside the development of a much larger Chinese bullion trading and settlement ecosystem.
-
The tactical setup remains supportive because ETF demand and options activity are returning while CTAs still have short exposure and gold volatility has not fully caught up with .
-
$4,400 to $4,500/oz is genuine resistance, but I would treat it as a decision zone rather than an automatic exit. Trimming makes sense; abandoning the trade does not unless gold is decisively rejected.
-
The bigger story is that gold may be moving beyond a simple Fed trade and beginning to price something much harder to resolve: fiscal credibility, sovereign balance sheets and the increasingly heavy debt load sitting underneath the global financial system.
Is Gold Asking a Different Question Now?
Gold spent the opening phase of the Iran conflict doing something that, at first glance, looked completely wrong. Missiles were flying, oil was ripping higher, geopolitical risk was suddenly being repriced across every asset class on the screen, yet gold kept struggling. The explanation was actually straightforward once you stopped looking at the war and started looking at the rates market. Gold was not trading geopolitical fear so much as the inflationary consequences of that fear. Every extra dollar in crude threatened to push inflation expectations higher, keep the Fed leaning in the wrong direction and leave real yields and the dollar doing the heavy lifting against bullion. Gold, normally the first asset handed the fire extinguisher when geopolitical smoke starts coming under the door, found itself being blamed for the heat.
That was the first phase of the Iran trade in a nutshell. Oil higher meant inflation risk higher, which meant Fed risk higher, which meant gold lower. Then when the immediate threat of escalation eased, crude backed away, the rates market relaxed and gold caught a bid again. For a while, the relationship was almost mechanical.
What makes the latest move interesting is that the machinery no longer seems to be working quite as cleanly.
Oil remains elevated. are hardly collapsing. The dollar has not fallen through the floor. Yet gold has begun pushing higher anyway. One or two sessions do not make a new regime, but markets rarely send engraved invitations when the regime starts changing. Usually, the first clue is much simpler: an asset stops falling on the news that is supposed to hurt it.
That is what has started to catch my attention.
Gold may no longer be asking only what oil means for the Fed. It may be starting to ask what all of this eventually means for the sovereign balance sheet.
That is a much larger question, and potentially a much more durable one.
Washington can massage the Treasury funding mix until the cows come home. It can lean harder on bills, hold coupon sizes steady, alter buybacks, smooth maturities and make quarterly financing announcements look as orderly as possible heading into the midterms. There will always be plenty of political noise around how Scott Bessent manages the Treasury book, and plenty of accusations that the funding profile is being dressed for the November window. I would leave the conspiracy theories to somebody else. Gold does not need them.
The arithmetic is uncomfortable enough.
The United States is still borrowing enormous amounts of money, interest costs keep climbing, and the stock of debt does not shrink because Treasury moved a few chairs around the dining room. Funding the deficit and convincing the world that funding the deficit indefinitely carries no consequences are two entirely different exercises.
Gold has always lived rather comfortably in the space between those two ideas.
That is also why the central-bank story deserves considerably more attention than the headline numbers gave it earlier this year. On the surface, official sovereign buying appeared to have cooled sharply. Dig underneath it, and the picture becomes considerably murkier. Goldman Sachs effectively had to rebuild parts of its central-bank demand model because conventional official statistics and trade data were failing to capture enough of the metal moving through London. The World Gold Council came to a similar conclusion using broader physical-market evidence. Once estimated unreported purchases were included, sovereign gold demand looked much healthier than the official disclosures suggested.
So the queue outside my local Bangkok gold shop may have looked shorter, but somebody was still emptying the shelves through the loading dock.
That is not a trivial distinction because central banks are not momentum tourists. They do not normally buy because a moving average crossed another moving average or because somebody on television drew an arrow pointing north. These are strategic allocations made against reserve composition, sanctions risk and the uncomfortable reality that foreign assets can cease being entirely yours when geopolitics turns ugly enough.
The freezing of Russian reserves in 2022 was not some obscure footnote in the history of sanctions. For reserve managers around the world, it was a demonstration. A dollar asset held offshore may be liquid, deep and backed by the world’s largest economy, but under extreme circumstances, it is still somebody else’s liability sitting within somebody else’s financial architecture.
Gold is nobody else’s liability.
China clearly understands the difference.
The People’s Bank of China continued adding to reserves through the first half of the year, with reported purchases accelerating again into June, while the physical market beneath those official numbers has been giving off its own signals. Shanghai Gold Exchange withdrawals jumped as prices corrected and the local supply chain restocked. Gold imports through Hong Kong remained substantial. Some of that was undoubtedly inventory building rather than an immediate explosion in jewellery or investment demand, but I think the market gets far too cute about that distinction.
Inventory has to be bought too.
If a tonne of gold is moved from the available market supply into a vault because somebody expects to need it for deeper trading, settlement, collateral or custody ecosystem, that tonne has still left the system available to everybody else. An ounce sitting in inventory ahead of a market launch is no less physical than an ounce sitting around somebody’s neck.
And that brings Hong Kong into the story.
China is clearly trying to build a larger Asian bullion architecture around Hong Kong, with deeper trading, clearing, custody and settlement links to the mainland. We should be careful not to overstate what that means on day one, but new financial infrastructure has a habit of creating demand before the final users even arrive. Build an airport and you need fuel, equipment, staff and inventory before the first passenger walks through the gate. Build a deeper bullion hub and somebody has to make sure there is actually bullion sitting behind it.
That may ultimately prove to be a much more important source of incremental Chinese demand than the market currently appreciates.
All of which would already be enough to keep the structural gold argument alive, but the tactical market is beginning to make the setup considerably more interesting. Retail money is returning to . Options activity is picking up, particularly further out the curve. Central-bank buying looks better than the official headline numbers implied. Yet systematic positioning is hardly screaming euphoric long, with CTAs still carrying enough short exposure to become buyers if the price keeps moving against them.
Then there is volatility.
Gold volatility moved enormously during the previous rally, and the market eventually became stuffed with overwriting and gamma supply as investors monetized the enormous premiums available in calls. That helped suppress vol as spot cooled. This time spot has started moving again, but volatility has not yet fully followed.
That is an interesting little mismatch.
The horse has started running before the bookmakers have completely changed the odds.
Gold has historically had a nasty habit of developing upside convexity once the market gets properly excited. Spot goes higher, calls get chased, dealers have to hedge, systematic shorts start covering, and volatility begins feeding back into the spot move. A fundamentally sensible rally can suddenly acquire the sort of self-reinforcing momentum that leaves everybody staring at yesterday’s option prices wondering why they looked so cheap.
We are not there yet.
That may be precisely why the setup is interesting.
Of course, none of this makes the technical picture disappear. Gold is heading directly into the $4,400 to $4,500/oz region, and there is plenty of scar tissue sitting there. Former support, the 200-day moving average and the broader downtrend all crowd into roughly the same neighbourhood. Anyone pretending that is not meaningful resistance is trading with one eye closed.
After the rally from the recent lows, taking some profit into that area makes perfectly good sense.
But taking profit and abandoning the trade are two very different things.
If this were still only the old gold trade — Fed cuts, falling yields, a weaker dollar and geopolitical fear — I would be far more inclined to treat $4,400 to $4,500/oz as a place to ring the register and go looking for the next horse. But the underlying argument now looks broader than that. Sovereign reserve diversification is still going on. Treasury financing needs remain enormous. China’s official purchases continue. Physical flows suggest that the Chinese gold ecosystem itself may be expanding. ETF demand is starting to wake up. CTAs still have shorts to cover. Volatility has not fully caught the move.
So for me, $4,400 to $4,500/oz is not simply resistance.
It is customs.
Gold has reached the border and now has to show its passport.
If yields keep climbing, the dollar strengthens, and gold gets thrown straight back from that zone, then trimming exposure was exactly the correct response. There is no medal awarded for turning a winning trade into a macro religion.
But if gold can live above $4,500/oz, particularly while oil remains firm and the Treasury market refuses to provide the traditional tailwind, the signal becomes far more interesting. Add rising volatility and the first meaningful CTA covering to that mix and the technical wall everyone is currently staring at could stop behaving like a ceiling and start behaving like fuel.
Resistance only works while somebody is prepared to sell it.
Once those sellers disappear, yesterday’s ceiling has a funny habit of becoming tomorrow’s floor.
That is really the part of this rally that has my attention. During the first stage of the Iran conflict, gold was staring at the fire alarm because every flare-up threatened another inflation impulse and another turn of the Fed screw. Now it appears to be looking past the alarm and studying the building inspector.
Perhaps the market is beginning to worry less about the next 25 basis points and a little more about the structure supporting the entire building.
And if gold really is transitioning from asking what will the do? to asking what ultimately backs all of this debt?, then this move may still have considerably more work to do.