Gimme Shelter: Why Gold Got Margin-Called by a War, What It Reveals About Silver, and Why We Remain Strongly Bullish

Gold suffered its worst week since 1983 — during an active war, with oil above $108 and Hormuz blockaded. The headlines say gold failed. We disagree. Here is our anatomy of the selloff, what it reveals about gold and silver’s true nature, and why our structural bull case is very much intact.

♪ War, children — it’s just a shot away, it’s just a shot away.

— The Rolling Stones, Gimme Shelter (1969)

Investors entering 2026 assumed that if war came, gold would be their shelter — the shot away from fiat chaos. In the short run, it hasn’t worked out that way. Understanding precisely why is what this report is about.

Let us begin with the uncomfortable truth that has dominated financial media this week. Gold hit an all-time high of $5,589 on January 28, 2026. By March 19 — with the United States and Israel at open war with Iran, the Strait of Hormuz blockaded, and the largest energy supply disruption in recorded history underway — gold had fallen to $4,551. That is an 18.6% decline in under two months, including the worst single week since 1983.

Silver, as is its nature, was considerably more dramatic: down 30.9% month-to-date, briefly plunging toward the 0.618 Fibonacci extension near $60.80. The metal we have long described as ‘gold on steroids’ has lived up to that characterisation — in both directions. We address silver in depth below, because its selloff mechanics and recovery potential differ importantly from gold’s.

We have spent nearly two decades studying gold cycles, monetary history, and investor psychology through our In Gold We Trust research. What we are witnessing right now is painful, counter-intuitive, and — in our considered view — entirely consistent with the anatomy of prior major cycle lows. We are not abandoning our structural thesis. We are explaining, with precision, why the short-term picture looks the way it does and what it implies for positioning.

Our Framework: A Golden Decade That Isn’t Over

In our In Gold We Trust 2025 report — The Big Long — we characterised the current period as the second half of a golden decade. Our base-case Incrementum Gold Model projects a gold price of $4,800 by 2030. In an explicitly stagflationary scenario — which we now believe is the base case, not a tail risk — we consider $8,900 plausible. Even at Friday’s close near $4,500, the market remains within our structural bull trajectory. The correction, however severe it feels in real time, is not a secular reversal. We have documented this pattern in IGWT editions through multiple cycles.

The annual gold performance table below provides necessary perspective. In USD, gold returned +26.3% in 2024 and +66.5% in 2025 — the strongest annual performance since the 1970s. Year-to-date in 2026, even after the war-induced selloff, gold remains +6.4% in USD and +8% in euros. Silver is still +90% on a twelve-month basis. The hysteria visible on financial social media is, to borrow a phrase we use frequently in our research, complaining from a very high altitude.

The structural pillars we have documented in every IGWT edition since 2022 remain intact. Central bank gold reserves reached 36,252 tonnes as of February 2026 — gold’s share of total reserves is now 18.2%, a 27-year high, through purchases that are price-independent and strategically motivated. These are sovereigns diversifying away from the dollar after the 2022 Russian reserve freeze proved that dollar-denominated assets can be weaponised. That lesson has not been unlearned. Financial investors, meanwhile, still allocate just 2% of portfolios to gold and precious metals. The reallocation pipeline is enormous, and it has barely begun.

Geopolitical Risk at Historic Highs — Yet Gold Fell

The Geopolitical Risk Index, as measured by Macrobond and HSBC, has spiked toward levels last seen after September 11 — dwarfing Russia-Ukraine 2022 and the Israel-Hamas war of 2023. By this measure, the Iran conflict is the most acute geopolitical shock in over two decades.

The common assumption — ‘buy gold on war’ — has always been more nuanced than the bumper sticker version. We have tracked this in our IGWT research for years. Goldman Sachs’ analysis of gold’s performance across 14 major geopolitical events since the Yom Kippur War shows a median return of just +2% with a hit ratio of 71%. What matters is the type of shock: deflationary fear drives gold higher. Inflationary shocks — especially energy-price-driven ones — can be deeply bearish near-term because they validate central bank hawkishness and strengthen the dollar. This war created the latter, not the former.

Goldman Sachs: Asset Returns Around Major Geopolitical Events (1973–2025)

Source: Datastream, Goldman Sachs Global Investment Research — Gold’s median return across 14 major crises since Yom Kippur: +2%. Oil’s median: +11%. The war-inflation combination is the most challenging near-term configuration for gold. It does not change our long-term thesis one iota.

Six Reasons Gold and Silver Sold Off: The Complete Anatomy

The selloff was not caused by one thing. Six distinct forces converged simultaneously. Any two or three of them would have produced meaningful pressure. All six together produced a historic washout. Let us work through each in turn.