Gold Is Falling Whether the War Escalates or Fades, and the Reason Is Macro

rose from approximately $5,040 to $5,423 in three sessions after the February 28 strikes on Iran. Within three weeks it had fallen to a cycle low of $4,090 — a 27% decline from its January 29 all-time high of $5,595 — even as Hormuz tanker transits collapsed from approximately 60 per day to approximately five and Brent crude exceeded $112 per barrel.

The war did not change sign. The market’s dominant transmission pathway did. What drove gold lower was not the absence of geopolitical risk but the inflation and rate dynamics that geopolitical risk produced — and the same mechanism is now running in reverse as ceasefire signals begin to dissolve the war premium.

Phase One: The Spike That Followed the Strike

Spot gold had been trading near $5,040 in the days before the February 28 strikes. Within three sessions it reached $5,423 — the classic safe-haven response to an acute geopolitical event, with institutional capital moving into non-sovereign assets as the scale of the military operation became clear. The move was precisely what the historical playbook predicted, and it lasted approximately 72 hours.

By March 3, spot had retreated to $5,085, retracing the majority of the initial gain. By mid-March it had fallen to a cycle low of approximately $4,090. This happened while Hormuz tanker transits were collapsing from approximately 60 per day to approximately five per day. Brent crude was exceeding $112 per barrel. The conflict was not fading. It was, by most observable measures, intensifying — and gold was falling 27% from its January 29 all-time high of $5,595.

The price action is not contradictory. Gold does not respond to war as a single variable. It responds to what war does to the two macro conditions that drive its cost of ownership: real yields and the dollar. When those two variables move against gold, the safe-haven bid loses to the carry cost — regardless of what is happening on the ground in the Middle East.

Technical Snapshot

FIGURE 1.  XAU/USD — Daily Technical Structure, Jan 29–Mar 26, 2026.  Three-panel chart: price with SMA20/SMA50, RSI(14), and MACD histogram. Event markers at the February 28 conflict onset and March 18 FOMC decision. Support and resistance zones shaded. Post-March 26 scenario arrows represent directional bias only. Sources: LBMA, ICE Futures, CME Group. For illustrative purposes only.

The price structure from February 28 to March 26 is a sustained directional decline, not a consolidation. Gold set progressively lower highs and lower lows across three consecutive weeks following the initial spike to approximately $5,423, a sequence that distinguishes trend deterioration from corrective noise. The 21-day SMA closed below the 50-day SMA on March 25, confirming a Bear Cross that formalises what the price structure had already communicated; both moving averages now slope downward and sit materially above current price at approximately $4,937 and $4,965, respectively, providing overhead supply at precisely the levels where prior recovery attempts have stalled. RSI near 31 is not oversold in the classical recovery sense — it is approaching the threshold associated with selling exhaustion, which is a different condition from a positive reversal signal. MACD histogram bars are narrowing on the negative side, indicating that selling momentum is losing force without yet reversing direction. The 200-day SMA near $4,100 is the structural reference that both the technical and macro arguments converge on: below it, the correction-within-an-uptrend framework requires revision.

Phase Two: Oil, Inflation, and the Rate Constraint

Brent crude reaching approximately $112 per barrel in the weeks after the Hormuz disruption activated the suppression mechanism for gold through two simultaneous channels. The first was inflation expectations. With tanker transits at approximately five per day and Gulf producers beginning to curtail output as land-based storage filled, the forward inflation signal shifted materially. Futures markets that had been pricing two Federal Reserve rate cuts in 2026 at the start of the year repriced to one, and the March 18 dot plot confirmed that view: seven of nineteen FOMC participants projected zero cuts for the full year, and the median PCE inflation forecast was revised upward 30 basis points to 2.7%.

The second channel was dollar strength. When the conflict began, global capital moved to the reserve-currency safety of the U.S. dollar at the same time it moved toward gold — but the dollar’s liquidity advantage sustained. The 10-year Treasury yield reached 4.39% on March 20, the highest level since July 2025. Gold yields nothing. When real yields rise and the dollar strengthens simultaneously, holding gold carries a rising opportunity cost, and the institutional capital that initially bought the safe-haven spike began rotating into yield-bearing dollar assets.

The FOMC hold on March 18 — 11 votes to 1 — did not change the immediate rate. What it changed was the implied path. A market that had been pricing the first cut in June 2026 was now pricing December at best. That single shift in rate-cut timing extended the period during which gold’s carry cost would remain elevated, and the price response was direct: from approximately $5,000 at the March 18 session to approximately $4,090 within four trading days.

Gold is being hit from both directions of the same event: when the war intensified, oil-driven inflation suppressed it by pushing rates higher; now, as ceasefire signals emerge, the war premium itself is dissolving. The mechanism is the same at every stage — macro repricing, not geopolitics.

Phase Three: The Same Mechanism Running in Reverse

On March 25, the Trump administration delivered a 15-point ceasefire proposal to Iranian authorities via Pakistan. Iranian leadership rejected the proposal — stating that no negotiations would begin until their conditions were met, including recognition of authority over the Strait of Hormuz — but the existence of a diplomatic channel was sufficient to move oil below $100 per barrel for the first time since mid-March. The transmission ran in the same direction it had run during the escalation phase, but in reverse: lower oil, softer forward inflation signal, marginal easing of rate expectations, reduced opportunity cost of holding gold.

The price response confirmed the mechanism. Spot gold recovered from the cycle low near $4,090 to approximately $4,560 on March 25 — a move of approximately 11% in roughly eight sessions, driven not by any new safe-haven demand but by the partial unwinding of the rate and dollar pressures that had suppressed the metal since early March. The safe-haven bid that the war itself generated was not returning. What was returning, briefly, was the removal of the headwind.

Iran’s rejection of the proposal, confirmed by state media the same evening, reversed the move. Gold fell below $4,500 the following session as ceasefire optimism faded and oil recovered. The pattern that produced the 27% decline from the January high compressed into a single week: the war premium formed on diplomatic signals and collapsed on diplomatic failure, with gold ending the week range-bound between approximately $4,490 and $4,530. This is the third phase of the same macro repricing dynamic — not a different story, not a new driver, but the identical chain of oil, inflation expectations, and rate assumptions operating in the opposite direction.

The structural medium-term case has not changed. Goldman Sachs holds a year-end 2026 target of $5,400, citing persistent central bank purchasing that has averaged above 1,000 tonnes annually since 2022. J.P. Morgan’s 2026 target of $6,000 is anchored to continued global de-dollarisation. Both of those structural tailwinds remain operative. What the oil shock introduced was a near-term timing constraint on the rate-relief that the medium-term case depends on. That constraint eases only when Hormuz traffic normalises durably and energy inflation stops feeding forward PCE estimates.

What to Watch

The Strait of Hormuz tanker count is the single most direct indicator of which macro phase is active. Daily transits fell from approximately 60 per day before the February 28 strikes to approximately five per day within nine days. A sustained recovery toward 20 to 25 transits per day would reduce the oil supply constraint and, through that, the forward inflation signal that has been holding the Federal Reserve on hold and keeping gold’s carry cost elevated. The sequence runs precisely in reverse of what drove the decline from $5,131 to $4,090 between early March and mid-March.

The April 10 CPI release is the second critical data point. The February and March CPI readings covered periods that partially precede the energy shock; April is the first print that will fully capture the Hormuz-driven pass-through. A reading above the Fed’s revised 2.7% PCE projection would extend the rate constraint into the second half of 2026. A reading at or below that level — whether because oil has stabilised near $98 or because demand destruction has begun to limit pass-through — would shift June cut probability from its current depressed level and ease the primary headwind on gold.

The medium-term structural case has not changed. De-dollarisation flows, persistent central bank demand, and the U.S. fiscal trajectory remain intact as multi-year tailwinds. What has changed is the near-term timing: the same conflict that appeared to justify owning gold produced the inflation and rate dynamics that made holding it expensive. When ceasefire signals emerged, the same mechanism removed the war premium rather than restoring the safe-haven bid. Gold did not misread the war. The market changed what it was pricing, and gold followed.

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 and analyst targets sourced from LBMA, ICE Futures, CME Group, FXStreet, TheStreet, FX Leaders, Advisor Perspectives, CNBC, and publicly available FOMC materials. All prices approximate as of March 26, 2026. Past price dynamics do not predict future behavior.