Gold reclaimed $4,000 after an in-line PCE print trimmed September hike odds from 68% to 63%. The structural bull case — central banks, real rates, a peak inflation narrative — is intact. But 298 tonnes of ETF gold sitting at a loss is a ceiling trad
Gold has had a brutal five weeks bottoming at $3,975/oz on Wednesday, its lowest print since November 2025, before clawing back above $4,000/oz on Thursday after May's PCE data came in exactly as expected.
The question every gold futures trader is asking right now: is that the low, or just a bounce on the way to something worse?
The honest answer is that both sides of this trade have a legitimate case — and the PCE print today handed the bulls their first real ammunition in weeks.
Why Thursday's PCE Matters More Than It Looks
Core PCE at 0.3% MoM and 3.4% YoY matched consensus to the decimal. That is not a soft print — inflation is still running nearly 70% above the Fed's 2% target. But what matters for gold is not the absolute level; it is what the data does to rate expectations.
September hike probability fell from 68% to 63% within minutes of the release. Treasury yields slid. The DXY pulled back from its 13-month highs. All three are mechanical tailwinds for a non-yielding asset that has been crushed by their mirror-image moves since the June 17 FOMC.
Critically, the in-line print removes the risk of an additional hawkish shock on top of Federal Reserve Chair Kevin Warsh's already-aggressive June pivot. As I argued in my oil disinflation piece published earlier today, May's PCE still reflects $107 WTI — the peak of the Iran war energy spike. June's reading will reflect $69 crude.
The Philadelphia Fed's forecasters project a 300 basis point swing in annualised CPI between Q2 and Q3, driven almost entirely by energy base effects. If that disinflation lands in the July PCE release, the September hike case weakens materially and gold's rate headwind may start to reverse.
The Structural Floor That Isn't Going Anywhere
Beneath the rate-driven noise, the structural bull case for gold is arguably stronger today than it was at the January high.
Global central banks added a net 290 tonnes in Q1 2026, the strongest start to a year on record. The People's Bank of China has increased its gold reserves for 20 consecutive months; China's net gold imports hit 317 tonnes in Q1 alone, nearly three times the prior quarter.
The WGC's 2026 survey found 84% of central banks expect gold to account for a larger share of global reserves over the next five years, while nearly 90% expect official gold holdings to increase over the next 12 months. This demand is price-insensitive, it does not capitulate on a hot jobs report or a hawkish dot plot.
Source: ING, World Gold Council
J.P. Morgan's commodities team is holding a $6,000 year-end target. The logic: real yields remain barely positive, the structural de-dollarisation trade is intact, and U.S. fiscal concerns — currently masked by the Iran war narrative will reassert themselves once the geopolitical dust settles.
On the flip side, ING Research have downgraded their forecasts again, now expecting gold to average $4,300/oz in the third quarter of 2026 and $4,600/oz in the fourth quarter, down from the previous forecasts of $4,850/oz and $5,000/oz, respectively.
The Bear Case: 298 Tonnes of Overhead
The ceiling is just as real as the floor. Goldsilver.com's analysis of ETF positioning data shows approximately 298 tonnes of gold currently held inside ETFs at a loss at current price levels.
Every bounce toward $4,065–$4,100 runs into that supply — holders who bought between $4,100 and $4,400 who are waiting to exit breakeven.
Gold Futures (GC1!) Four-Hour Chart, June 25, 2026
Source: TradingView
The honest GC setup today: the PCE no-surprise removes the immediate downside risk and gives bulls a credible bounce to work with.
The $3,975 Wednesday low is the line — a weekly close below it reopens the path toward $3,800.
To the upside, $4,065–$4,100 is the first meaningful resistance; clearing $4,258 with volume is what shifts the narrative from dead-cat bounce to trend reversal.