Key Takeaways
- Gold’s collapse: Gold fell 23%+ from its March 2025 all-time high of ~$3,167/oz to approximately $2,430/oz by mid-March 2026 — despite an active Middle East war.
- Saudi holdings: Saudi Arabia holds approximately 323 tonnes (~$41B at peak) in official gold reserves — unchanged per latest IMF data, with no confirmed sales.
- UAE and Kuwait: UAE holds ~11 tonnes, Kuwait ~79 tonnes — both relatively minor positions. Neither central bank has disclosed sales.
- The real sellers: IMF data shows net EM central bank gold buying slowed sharply in Q1 2026; some Eastern European and Latin American central banks reduced positions.
- SWF wildcard: Gulf sovereign wealth funds (ADIA, Mubadala, PIF) may hold gold indirectly via ETFs or forwards — these positions are not disclosed in IMF data.
Gold’s behavior in March 2026 has confounded every textbook prediction. An active regional war, a record-high US deficit, a collapsing Egyptian pound, and a Hormuz closure that rattled energy markets for weeks — and yet gold fell from its all-time high of approximately $3,167 per ounce (March 2025) to below $2,430 by mid-March 2026, a decline of more than 23%. The internet’s leading theory: Gulf sovereign wealth funds are dumping gold to fund war-related expenditures.
We investigated. The reality is more nuanced — and more important for investors in GLD, IAU, or any gold-linked instrument.
What Do Gulf States Actually Hold in Gold?
The first step is separating official central bank reserves — which are disclosed to the IMF — from sovereign wealth fund holdings, which are largely opaque.
Saudi Arabia is the largest Gulf gold holder. The Saudi Arabian Monetary Authority (SAMA) reported 323.07 tonnes of gold in its latest IMF COFER submission, valued at approximately $41 billion at peak 2025 prices. This figure has been essentially static since 2010, when Saudi Arabia made a one-time disclosure of a large gold position it had been holding undisclosed. There is no IMF-reported evidence of Saudi gold sales in 2025 or early 2026.
Kuwait holds approximately 79 tonnes — a more modest but still meaningful position relative to its reserve base. The UAE reports only approximately 11 tonnes through its central bank, reflecting the fact that Abu Dhabi’s investment assets are channeled primarily through sovereign wealth funds rather than the central bank.
Qatar holds approximately 44 tonnes. Bahrain and Oman hold negligible quantities. Across all six GCC central banks, total disclosed gold is approximately 460–470 tonnes — roughly 1.5% of global central bank gold holdings.
For additional context on Gulf financial positions, see our analysis of Saudi Arabia’s TASI market disconnect from oil prices.
Why Did Gold Fall 23% During an Active War?
The counterintuitive decline requires understanding what actually drove gold’s 2024–2025 surge. Gold’s all-time high was not primarily a war-premium trade — it was driven by central bank accumulation (particularly China, India, and Turkey buying at scale), de-dollarization narratives among EM central banks, and retail demand in Asia. All three drivers reversed in late 2025 and early 2026.
China’s People’s Bank paused gold purchases in Q4 2025 after accumulating more than 300 tonnes over three years. India’s central bank became a net seller in January 2026 as the rupee came under pressure and the RBI needed liquidity. Turkish retail gold demand collapsed as the lira stabilized and domestic interest rates remained attractive in real terms.
Simultaneously, the Iran war created an unexpected dollar-positive shock. Energy-importing nations scrambled for dollars to pay elevated oil bills; the DXY (Dollar Index) strengthened materially, which historically pressures gold. US 10-year real yields also rose as defense spending expectations pushed Treasury supply projections higher — a headwind for non-yielding gold.
Our earlier gold analysis covers this dynamic in detail: Why Gold Is Falling Despite War Escalation — March 2026.
The Sovereign Wealth Fund Question: What We Don’t Know
Here is where the mystery gets genuinely interesting — and where the “Gulf selling gold” narrative has a kernel of truth, even if the official reserve data doesn’t confirm it.
Gulf sovereign wealth funds — ADIA ($993B AUM), Mubadala ($302B), PIF (~$700B), QIA (~$450B) — do not disclose asset-level holdings. They are not required to report to the IMF. It is entirely plausible that these funds hold gold indirectly via:
- Physically-backed ETF positions (GLD, IAU, iShares Physical Gold)
- Gold futures and forward contracts via their commodity desks
- Allocated gold accounts at LBMA-member bullion banks
- Equity stakes in gold mining companies
If any of these funds reduced gold exposure in Q1 2026 — to raise liquidity, rebalance portfolios after gold’s 2025 gains, or fund commitments elsewhere — that selling would not appear in IMF central bank reserve data. It would, however, appear in the gold market.
LBMA vault data shows net outflows from London-registered gold vaults of approximately 45 tonnes in January–February 2026, a level elevated relative to the 2023–2024 average. The destination of those flows is not publicly disclosed. This is the factual basis for the “Gulf selling” speculation — real metal is moving, but we cannot confirm the seller’s identity.
What This Means for US Investors
For US investors holding GLD or IAU, the Gulf selling question matters because it changes the floor calculus. If Gulf SWFs were significant buyers during gold’s 2024–2025 run — and there is circumstantial evidence they were, given their stated de-dollarization rhetoric — then any sustained reduction in their exposure removes a structural buyer. The key data point to watch: LBMA vault outflows. If net London vault outflows exceed 100 tonnes in Q1 2026, it would suggest institutional liquidation at a scale consistent with the “large sovereign seller” hypothesis. Current data (45 tonnes through February) is elevated but not alarming. A secondary signal: GLD/IAU premium/discount to NAV. Persistent discounts would suggest institutional redemption pressure rather than retail panic. As of mid-March 2026, GLD trades near NAV — consistent with orderly institutional repositioning rather than a distress sale.
What Does IMF Data Actually Show for Q1 2026?
The IMF’s COFER (Currency Composition of Official Foreign Exchange Reserves) data through Q4 2025 shows global central bank gold purchases slowed to approximately 290 tonnes for full-year 2025, down from the record 1,082 tonnes in 2022 and 1,037 tonnes in 2023. The deceleration was the primary driver of gold’s loss of momentum heading into 2026.
Notably, no GCC central bank shows a reduction in gold holdings in any available IMF submission. Saudi Arabia’s 323 tonnes has been flat for 15 years. This strongly suggests that if Gulf entities are selling gold, it is through SWF channels (undisclosed) rather than central bank channels (disclosed). The two channels serve different purposes: central bank gold is a reserve asset and strategic anchor; SWF gold is a return-seeking investment that can be traded freely.
For the gold price outlook, see our forecast piece: Gold Price Forecast April 2026: Safe Haven Dynamics.
Could Gulf States Use Gold Sales to Fund War Costs?
Theoretically yes; practically, it is unlikely to be necessary. Saudi Arabia’s fiscal breakeven oil price is approximately $76/barrel; at current Brent prices above $100, Riyadh is running a substantial fiscal surplus. The UAE and Kuwait are similarly cushioned. Qatar’s LNG revenues, despite the Doha terminal strike in February, remain strong due to long-term contracts with price escalators.
The scenario where Gulf states are forced to sell gold to fund operations requires oil prices below $65 for an extended period — the opposite of the current environment. The more plausible scenario is portfolio rebalancing: after gold’s 38% gain from 2023 to 2025 peak, SWFs with disciplined rebalancing policies would mechanically reduce gold exposure as it grew beyond target allocations.
Related: Iran Strikes Qatar LNG Terminal — Brent Surges to $108.
Frequently Asked Questions
Are Gulf states selling their gold reserves in March 2026?
Official IMF data shows no reduction in GCC central bank gold holdings. Saudi Arabia’s 323 tonnes, Kuwait’s 79 tonnes, and Qatar’s 44 tonnes are all reported stable. However, Gulf sovereign wealth funds (ADIA, Mubadala, PIF, QIA) do not disclose asset-level holdings. LBMA vault outflows of ~45 tonnes in Jan-Feb 2026 are elevated, raising the possibility of undisclosed SWF repositioning — but this cannot be confirmed.
Why is gold falling during an active Middle East war?
Gold’s decline reflects the reversal of its 2024–2025 bull drivers: Chinese central bank buying paused, Indian RBI became a net seller, and Turkish retail demand fell. Simultaneously, the war created dollar strength (energy importers buying USD) and pushed real US yields higher — both headwinds for gold. The war premium that textbooks predict failed to materialize because gold’s prior rally had already priced in significant geopolitical risk.
How much gold do Gulf Arab states hold collectively?
GCC central banks collectively hold approximately 460–470 tonnes of gold in disclosed reserves: Saudi Arabia 323t, Kuwait 79t, Qatar 44t, UAE 11t, with small amounts from Bahrain and Oman. This represents roughly 1.5% of global central bank gold holdings. Sovereign wealth fund holdings are additional and undisclosed.
What happens to GLD and IAU if Gulf SWFs are selling?
If Gulf SWFs are reducing GLD/IAU exposure specifically, fund outflows would drive share redemptions and metal sales. As of mid-March 2026, GLD trades near NAV — suggesting orderly repositioning rather than distress. Watch for sustained GLD discounts to NAV or rapid AUM declines as the key signal. A confirmed large-scale institutional withdrawal would remove a structural buyer and could extend gold’s correction toward the $2,200–$2,300 range.
Does Saudi Arabia need to sell gold to fund its war budget?
No. Saudi Arabia’s fiscal breakeven is approximately $76/barrel; Brent above $100 generates a large fiscal surplus. Riyadh can fund elevated military and diplomatic expenditures from oil revenues without touching its 323-tonne gold reserve. Gold sales under current oil market conditions would be unusual and would likely signal a significant deterioration in the fiscal outlook — which analysts do not currently project.
Data sourced from IMF COFER database, World Gold Council, LBMA vault statistics, and Gulf central bank annual reports as of March 2026. This article is for informational purposes only and does not constitute investment advice.
