Why Central Banks Won't Stop Buying Gold Even as Rates Stay High: The Long Game Beyond Real Yields
My phone has been buzzing all week with the same question: how is gold sitting at $4,500 while real yields are this high?
High real yields should crush gold. They raise the opportunity cost of holding a zero-yield asset, and that relationship was gospel for anyone who traded through the 2013 taper tantrum or the 2018 rate cycle. For a decade, it held up.
But central banks are buying at record pace anyway. The World Gold Council keeps reporting central bank gold buying above 1,000 tonnes a year, three years running, and the buying continues even as forecasts get cut and real yields stay elevated.
Most traders read this as a contradiction. I read it as a structural shift. Central banks aren't confused about real yields. They just don't care about them.
The Real Yield Model Is Breaking
The real yield trade was beautiful while it lasted. Sell gold when 10-year TIPS yields climb, buy it when they fall. Clean, simple, profitable. I ran it myself in the early 2010s, and the correlation was so tight you could set your watch to it.
Then 2022 broke it. Russian reserves got frozen, and every non-Western central bank received the same message: your dollar reserves are only as safe as your relationship with Washington. That was the turning point.
Since then, banks like OCBC have cut their gold price forecast because higher real yields weigh on precious metals. That's a defensible call if you're managing a macro book with a 12-month horizon. But it keeps missing the bigger picture. The forecasts get cut, and price keeps finding bids. Why? Because the forecasters are projecting their own framework onto buyers who don't share it.
The result is a gold market where real yields explain less and less of the price action. The old correlation is still visible on a chart, but it's no longer driving the marginal buyer. Central bank demand is.
I remember the exact feeling of watching that correlation start to slip in the summer of 2022. I kept shorting gold into strength because the real yield model told me it had to reverse. I got stopped out, sat on my hands, and watched the market grind higher for months. The tuition was painful, but the lesson was clear. The model wasn't wrong historically. It was wrong structurally. The buyer mix had changed.
Watching the Structural Bid in the Tape
I spend most of my time on D1 and H4 structure, and I've been watching something this month that keeps confirming what I just described.
In mid-July, gold sold off into the $3,960 to $4,000 zone. By the old model, that should have snowballed. Real yields were still elevated, the dollar wasn't collapsing, and momentum was clearly to the downside. But the selling died there. Every attempt to break lower got absorbed, and price is back above $4,500 now.
That's what a structural bid looks like. It doesn't announce itself with a headline. It shows up as a series of failed breakdowns. The washout gets bought. The retracement is shallower than it should be. The model says "this should drop," and price says "not yet."
When I see that pattern while ETF flows are flat, I don't scratch my head. I look for the buyer who doesn't care about the weekly chart. That buyer is a reserve manager in a country that would rather hold physical bullion than a T-bill.
I could be wrong about this. The model could reassert itself, and gold could break down the way the screens say it should. But after a decade of screen time, I've learned to trust failed breakdowns over failed forecasts. The tape is telling me something, and the tape doesn't care about my opinions.
This is also why the recent "gold is overbought" warnings from places like FOREX.com and FXEmpire read differently to me than they do to most people. They're watching momentum and positioning on H1 and H4, which is fine for a short-term trade. But they're not measuring the central bank bid that sits underneath the chart. That's not a criticism. It's a different time frame. You just need to know which one you're trading.
Central Banks Don't Trade Yields, They Trade Survival
Ask yourself this. Is a reserve manager going to explain to his finance minister that he kept 30% of reserves in dollars because the real yield on 10-year TIPS was attractive? No. He's going to explain why the reserves survived a sanctions event, a currency crisis, or a geopolitical rupture.
That's a completely different optimization problem.
For a macro hedge fund, gold is a tactical asset. You buy it when the real yield picture supports it, and you sell it when it doesn't. For a central bank, gold is insurance. It pays nothing, costs money to store, and produces no cash flow. And that's exactly the point. The dollar has become a political instrument. The Treasury market, for all its depth, has become a potential weapon. When you're on the wrong side of that weapon, your reserves are gone. Gold is the one reserve asset that doesn't have a counterparty that can freeze your account.
This is the currency debasement hedge argument, and it's not a fringe theory. It's portfolio diversification at the sovereign level. The IMF's own data shows the dollar's share of global reserves declining over the past two decades, and central bank gold buying is the mirror image of that trend.
Think about what that means for the standard gold market analysis. The opportunity cost framework assumes the alternative to gold is a safe, liquid, income-bearing asset. But for a growing number of countries, the dollar is none of those things. It's a liability with political strings attached. Real yields don't matter if you can't access the asset when you need it most.
That's why central banks keep buying through rate hikes, through yield spikes, through every environment where the textbooks say they should be sellers. They're not buying gold as an investment. They're buying it as a hedge against a world order that's becoming less reliable by the quarter. Call it a long-term investment strategy with a geopolitical twist.
Two Games, Two Time Horizons
Here's where I think most analysts get confused. They observe that central bank buying doesn't respond to real yields, and they conclude that central banks are being irrational. But they're not irrational. They're playing a different game with a different clock.
A trader thinks in sessions and weeks. A macro hedge fund thinks in quarters and years. A central bank reserve manager thinks in decades and generations. The Fed could cut rates tomorrow or hike again next month, and it barely registers in that decision framework. What registers is the trajectory of US fiscal policy, the reliability of the dollar as a neutral reserve currency, and the probability that sanctions get broader rather than narrower.
From that vantage point, gold at $4,500 is not expensive. It's cheap insurance against a fiscal and geopolitical tail that could make today's real yields irrelevant.
This is the point I keep returning to when I hear the "gold is in a bubble" calls. A bubble is when price detaches from fundamentals. But the fundamentals have changed. When you reprice gold as a strategic reserve asset instead of a yield-sensitive commodity, the levels make a lot more sense. The record central bank gold buying isn't a side story. It's the main story.
You can see this split in the flows. Gold ETF traders have been net sellers during parts of this rally, while central banks keep accumulating. The two biggest pools of demand are moving in opposite directions, and price keeps grinding up because the central bank bid absorbs everything the ETF sellers throw at it. If you're waiting for high yields to finally break gold's back, you're sitting in a yield trap. The carry on a 10-year TIPS looks great until the asset itself becomes politically unusable.
Anyway, that's the theory. Here's what I actually do with it on my own screen.
What This Means for the Gold Price Outlook
Real yields still matter on a tactical timeframe. When TIPS yields spike, gold will still wobble, and I respect that. I trade that. But the structural bid changes the shape of the pullbacks. Dips get bought sooner and don't go as deep. Retracements find support at 38.2% or 61.8% Fibonacci levels that would have been violated in a pure yield-driven regime.
The market now runs on two layers. The top layer is momentum and yield expectations, traded by ETFs, macro funds, and people like me. The bottom layer is central bank accumulation: slow, steady, and price-insensitive. When the layers align, you get explosive moves. When they conflict, you get grinding ranges that confuse the consensus.
I'm watching the D1 structure above $4,500 right now. If we break this level with central bank buying still running above 1,000 tonnes a year, the next leg up is a question of when, not if. If we fail here, the pullback probably stays shallow, and I'll be looking to buy the retracement instead of chasing it.
The uncomfortable truth for anyone still trading the old model: you can keep being right about real yields and keep losing money in gold. The marginal buyer doesn't care about your model. You can call that unfair, but the market doesn't owe you a correction that fits your thesis.
If you missed this rally, I get it. The pain for most traders I talk to is not the direction, it's the entry. Waiting for a deep correction might mean waiting forever, because the structural bid keeps shortening every pullback. And if you're a retirement planner building a long-term allocation, the strategic case writes itself: you want something that doesn't depend on the Fed's next move. Central banks are already doing it for you.
Gold at $4,500 Is Not a Bubble. It's a Signal.
I've been watching markets long enough to recognize a bubble. From everything I can see, this is not one. This is a repricing of gold's role in the global monetary system, driven by institutions that don't care about the next Fed decision.
Record central bank gold buying in a high interest rate environment is not a paradox. It's a statement. These institutions are hedging against a world where the dollar is a weapon, where fiscal deficits follow an unsustainable path, and where the "safest asset in the world" is not the safest one for them.
So here's my simple conclusion. Don't fight the central banks. When gold pulls back on a real yield spike, buy the dip. When it breaks out, don't short the extension just because the old model says you should. The model that worked for a decade is now the second most important thing in the room.
The question that keeps me up at night is different. At what price do central banks decide they've built enough reserves? When does the structural bid start to fade? I don't know the answer, and neither does anyone who tells you they do. What I do know is that the game changed in 2022, and a lot of traders are still playing the old one.
What do you think the endgame looks like when the biggest buyers in gold have no yield target and no exit plan? That's the question worth asking before you short the next record high.