Gold Is Falling Because of the War, Not Despite It

is supposed to thrive when geopolitical risk escalates. Strait of Hormuz traffic has been severely disrupted, oil has risen above $100, and the Middle East conflict shows no near-term resolution. Instead, gold has fallen more than 20 percent from its January intraday record and recorded its worst weekly performance in over six years.

The contradiction resolves when the transmission mechanism is understood. The same war that should support gold is driving oil higher, lifting inflation expectations, repricing the Federal Reserve’s policy path, and raising the real cost of holding a non-yielding asset. Gold is not failing as a safe haven. It is reacting to rates, which are reacting to oil, which is responding to the war.

The Setup: A Crowded Trade at Record Valuation

Gold reached an intraday high of $5,595 per ounce on January 29, 2026 — an all-time record on spot — before closing the month at approximately $4,894 as the initial speculative surge reversed. The move to that intraday peak reflected a specific macro configuration: a weakening dollar, multiple Federal Reserve rate cuts priced into forward markets, nine consecutive months of North American ETF inflows, and record central bank buying running at approximately 60 tonnes per month. In that environment, gold was priced to receive monetary easing. It was not priced for a sustained energy shock that would make easing more difficult.

The pre-conflict positioning was structurally elevated. Global gold ETF assets under management had doubled to an all-time high of $559 billion by year-end 2025. Physical holdings reached 4,025 tonnes, the highest on record. Western exchange-traded funds had absorbed roughly 500 tonnes since early 2025. High-net-worth investors and family offices held physical bars as a hedge against what Goldman Sachs described as the structural debasement trade — a combination of fiscal deficit concerns, dollar reserve erosion, and central bank credibility risk. That positioning required no conflict to sustain itself. It required only that the macro conditions supporting it, specifically easier monetary policy and a weaker dollar, remained directionally intact.

When U.S. and Israeli forces struck Iran on February 28, the initial reaction was textbook: spot gold spiked to $5,738 intraday on March 3, safe-haven demand intersecting with the geopolitical premium. That spike lasted approximately three sessions. What followed was not consolidation. It was a systematic unwinding of a crowded positioning structure into a macro environment that had fundamentally shifted.

Technical Snapshot

FIGURE 1  ·  XAU/USD Daily Price, Volume, and RSI(14)  ·  January – March 23, 2026. Panel 1: with ATH annotation at $5,595 (January 29), conflict-onset spike (February 28), and post-FOMC selloff from March 18 onward. Panel 2: Volume profile with elevated bars at ATH, GLD outflow day, and FOMC crash sessions. Panel 3: RSI(14) showing the progression from overbought extremes above 80 in late January through the current deeply oversold reading near 14. Sources: LBMA, COMEX. For illustrative purposes only.

The chart structure describes a distribution phase followed by a disorderly decline, not a normal pullback within an intact uptrend. Gold peaked with RSI readings in overbought territory in late January — consistent with speculative climax conditions rather than a measured accumulation advance. The $5,000 level held as support through mid-March before the FOMC’s hawkish hold on March 18 broke through it on elevated volume, a technically significant level given its role as a prior psychological ceiling and accumulation zone during the late-2025 rally. The $4,200 zone, where the 200-day moving average now resides, is the next structurally meaningful reference and the boundary between the current bull cycle’s demand base and the 2025 breakout range. RSI on the daily chart has extended deeply into oversold territory, a reading more consistent with forced liquidation than with orderly selling, while MACD histogram remains negative with no narrowing visible — neither indicator shows evidence of a momentum reversal. The most important observation from the price-yield relationship after March 18 is directional: in a conventional risk-off episode, bond yields fall as capital rotates into safe-haven Treasuries, and gold rises in parallel. Here both gold and yields moved in opposite directions from that script — yields rose and gold fell — which confirms that the decline is rate-driven rather than driven by growth fear or recession risk.

The Mechanism: Why This War Is Bearish for Gold

The standard safe-haven framework assumes geopolitical risk drives capital into gold because it creates recession fear, deflation risk, or policy easing. The current episode breaks that assumption at each step. The Iran conflict did not generate recession expectations. It generated inflation expectations. Brent crude above $105 represents a supply-side price shock that transmits directly into CPI. The February CPI, released March 11, printed in line with consensus at 2.4 percent year-over-year — but that reading was measured before the February 28 strikes. March and April CPI will be the first to capture the energy pass-through, and if oil prices remain near current levels, the forward inflation signal is more adverse than February’s in-line print indicated.

The Federal Reserve’s March 18 decision confirmed the policy consequence. The FOMC held rates at 3.50 to 3.75 percent and revised the dot plot median projection for year-end 2026 from 2.9 percent to 3.4 percent — effectively reducing the expected number of cuts from two to one. PCE inflation was revised upward to 2.7 percent. Powell’s press conference was interpreted as measured rather than aggressively hawkish, but in an environment where energy prices are rising and the prior expectation had been for easing, even a neutral communication is functionally more restrictive than its January equivalent.

The provides the direct transmission channel. It closed at 4.39 percent on March 20, the highest level since July 2025 and up from approximately 4.10 percent before the conflict began. Real yields, the opportunity cost of holding a non-yielding asset like gold, rose in parallel. Standard discounted-value frameworks for gold suggest that a sustained rise in real yields tends to compress the metal’s relative attractiveness, independently of any change in geopolitical conditions. The oil shock did not create safe-haven demand for gold. It created conditions under which holding gold became more expensive relative to the alternatives it competes with for portfolio space.

The war made oil expensive, oil made inflation worse, inflation made rate cuts harder, and the cost of holding gold rose — the metal is trading exactly as the transmission mechanism requires.

The Amplifier: How Positioning Extended the Move

The magnitude of gold’s decline reflects not only the macro shift but the structure of positioning that preceded it. Gold had attracted the kind of crowded, cross-institutional allocation that creates both upside momentum on the way in and forced liquidation on the way out. When the macro narrative reversed, three sequential flows compounded the move.

The first was margin-driven. As gold fell from the January peak, leveraged futures positions reached margin thresholds. Forced unwinding of those positions created selling pressure independent of any fundamental view change. The recorded a single-day outflow of $2.91 billion on March 4 — the largest since 2016 — consistent with institutional desks using the ETF as a liquidity source to meet margin calls elsewhere in their books. When the largest and most liquid gold instrument becomes the funding vehicle for other positions, the selling has nothing to do with gold’s structural case and everything to do with capital availability.

The second was allocation-based. Institutional investors managing multi-asset portfolios had built meaningful gold overweights through late 2025 and early 2026, partly as a dollar hedge and partly as a rate-cut beneficiary. When the Fed’s dot plot revised cuts out of the near-term calendar, the thesis underpinning the overweight weakened. Reducing gold and rotating into short-duration Treasuries — now yielding 4.39 percent — became a straightforward rebalancing trade. The strongest year on record for gold ETF inflows in 2025 meant there was more to sell.

The third was dollar-competitive. Safe-haven demand in a geopolitical shock historically splits between gold and the dollar, with dollar liquidity generally prevailing in the immediate term. As dollar strength became sustained post-FOMC, non-dollar buyers faced rising local-currency gold prices even at declining dollar-denominated levels. That dynamic suppresses demand from precisely the buyers, European and Asian central banks and retail investors, whose structural purchasing has underpinned the multi-year bull market.

Scenarios: What the Next Move Requires

What to Watch

Three variables determine whether the current decline has further to run or whether the

conditions for a stabilisation are assembling. The first is the Strait of Hormuz traffic count. According to tanker tracking data, daily commercial transits fell sharply from pre-conflict levels of approximately 60 per day in the weeks after the February 28 strikes, with war-risk insurance costs surging and many operators voluntarily suspending passage. A credible resumption of commercial flow would reduce the oil-inflation signal that is driving rate expectations higher. It would not immediately reverse the FOMC’s dot plot revision, but it would remove the forward energy shock from the March and April CPI reads. Every week of severely disrupted Strait traffic extends the period over which oil prices feed into measured inflation, making the Fed’s easing timeline harder to advance.

The second is the April 10 March CPI release. February’s in-line print delayed the market’s reckoning with the oil shock, because the reference period preceded the conflict. The March print will be the first data point to capture energy pass-through from February 28 onward. A reading above 2.7 percent year-over-year would validate the Fed’s dot plot revision, remove June from any serious discussion of a rate cut, and extend the rate headwind for gold through at least the June FOMC meeting. A reading at or below 2.4 percent — possible only if energy pass-through is slower than expected — would create the first macro condition for gold stabilisation since the conflict began.

The third is the 200-day simple moving average, currently near $4,200. That level represents the accumulated demand from late 2025 and the structural foundation of the current bull cycle. In prior corrections of comparable magnitude, the 200-day has consistently attracted central bank and long-horizon institutional buying. Whether that structural bid materialises at $4,200, or whether the combination of rate headwinds and positioning unwind pushes gold through it, is the technical and fundamental question that will define the next major directional phase. A closing break below $4,200 on elevated volume would represent a change in the cycle, not merely a correction within it. Until that break occurs, the weight of evidence — central bank buying, fiscal deficit concerns, dollar reserve erosion — still supports the medium-term structural case, even as the near-term rate environment remains unfavorable.

***

DISCLAIMER  This article is for informational and analytical purposes only and does not constitute investment advice or a solicitation to buy or sell any security. Price data sourced from LBMA, COMEX, CME FedWatch, World Gold Council, SPDR State Street Global Advisors, and publicly available FOMC materials as of March 23, 2026. Past episode dynamics do not predict future price behavior. Always consult a licensed financial advisor before making investment decisions.