Gold's ETF story is too Western
Saturday, Nov. 14, 2026 — A fortnight of GLD selling has been turned into a verdict on gold, and that is wrong. October's fund data show the weakness is Western, and more specifically US-led, not universal.

A fortnight of GLD selling has been turned into a verdict on gold. We think that is wrong. October's fund data show no global ETF flight at all: physically-backed gold ETFs added US$2.7bn and 19 tonnes, with North America the only region in outflow, while Europe and Asia both added money and India kept buying. Gold is under pressure, but the weakness looks Western, and more specifically US-led, not universal.
The prevailing interpretation says much the same at first glance: North American outflows were offset by stronger inflows elsewhere, so the global ETF bid is still intact. That view is only partly right. It is right on the geography. It is too casual about market structure. GLD is not the whole gold ETF universe, but it is still big enough to drive the marginal price. So Western selling can do real damage without proving that investors everywhere are leaving gold.
GLD is too big to ignore, but too small to explain everything
SPDR Gold Shares held 1,040.7t on Nov. 12, according to SPDR Gold Shares data compiled by GoldSquawk.
Compare that with the World Gold Council's 4,275t of global physically-backed ETF holdings at end-October, and GLD alone accounts for roughly 24% of the total, using 1,040.7 divided by 4,275. That is why traders stare at it.
But the price move has been far larger than the metal leaving the fund. Over four weeks, GLD holdings fell from 1,054t on Oct. 15 to 1,041t on Nov. 12, down 13t or 1.2%, while COMEX gold fell from $4,170.10 to $3,885.30, down 6.8%, according to SPDR Gold Shares. Over 13 weeks, GLD holdings rose from 1,023t to 1,041t, up 18t or 1.8%, even as COMEX gold dropped 12.1% from $4,420.40 to $3,885.30. That is not a clean ETF-led liquidation. It is a market where price has fallen much faster than the biggest fund has been emptied.
The same point shows up in the shorter-term flow data. SPDR Gold Shares data show GLD and GLDM took in 3.2t in the latest week to Nov. 12, even though they lost 6.3t over four weeks and gained 28.0t over 26 weeks. A market can clearly sell off while the largest Western products are still taking some money.
October's map was regional, not global
The World Gold Council's October map is unambiguous. Global physically-backed gold ETFs added US$2.7bn in October and holdings rose 19t to 4,275t. North America lost US$512mn and was the only region with outflows. Europe added US$1.8bn, about 3.5 times the North American loss, using 1.8bn divided by 512mn, while Asia added US$1.3bn, about 2.5 times the North American loss. That is not a broad retreat from gold. It is a split market.
India strengthens the point. AMFI reported October gold ETF net inflows of ₹3,474.03 crore, taking assets to ₹1,79,894.44 crore. The World Gold Council also said India saw sustained gold ETF buying in October. If readers are being asked to treat US-listed fund weakness as the whole story, these numbers say no.
This is where the comforting version of the global-inflow argument goes too far. Yes, the global tally was positive. But that does not mean all parts of the ETF complex are pulling together, or that Europe and Asia can simply cancel out US price-setting power day by day. October showed resilience in aggregate, not harmony.
The tape is following US fear, not the whole fund complex
The easy explanation for gold's break is now familiar: Fed rate risks rose and GLD bled. There is truth in that. Gold fell from $4,009.80 before the Nov. 11 headline to $3,857.50 at the Nov. 12 low, a drop of $152.30 or 3.8%, using GoldSquawk price data. COMEX futures then ended Nov. 12 down $113.60, or 2.84%, and Nov. 13 down just $0.10, according to GoldSquawk settlement data.
But the macro evidence at the end of the week did not show a fresh inflation shock. US initial claims were 200K against a 194K forecast on Nov. 12. US PPI rose 0.3% in October against a 0.4% forecast on Nov. 13, while core PPI was 0.3%, in line with forecast. By then the damage was already done. So we should be wary of any story that pretends a globally bearish ETF move confirmed the price break. The October flow map says the opposite.
A broader point follows. The World Gold Council said Q3 total gold demand was 1,369t, down 1% year on year, while the value of demand rose 22% to US$188bn. That is not the backdrop of a market from which capital has simply walked away.
What this means beyond this week
If we are right, US ETF outflows can cap rallies without confirming a durable bear market. That is the practical implication. Traders who read GLD as a proxy for all ETF demand will overstate the bearish signal whenever Europe and Asia are still adding metal.
The strongest evidence against us is also obvious. GLD is huge at 1,040.7t. North America was the only region in outflow in October. If those outflows deepen and the next World Gold Council update shows global totals turning negative, then the regional-rotation argument weakens quickly. Likewise, if SPDR's latest 3.2t weekly inflow reverses hard after next week's US retail sales on Nov. 17 or the FOMC minutes on Nov. 18, then Western selling may be spreading rather than staying local.
For now, though, the cleaner reading is not that investors have abandoned gold. It is that the US is dragging the price lower while the rest of the ETF world has not followed. That is a very different signal, and gold readers should treat it that way.
Key data
| Measure | Value | Source |
|---|---|---|
| GLD holdings | 1,040.7 t | SPDR Gold Shares, via GoldSquawk data |
Primary sources
- World Gold Council · Europe remains in driver's seat
- World Gold Council · Central bank buying and ETF inflows support gold demand in Q3
- GoldSquawk data: Gold ETF holdings (GLD) (SPDR Gold Shares); Gold ETF holdings by fund (Fund issuers); Gold ETF flows (SPDR Gold Shares)
General market commentary only. Not financial advice.