Gold's performance at the start of this year has been truly unusual. After reaching a high of $5,589 in January, it fell to around $4,100 in less than two months, with the monthly decline setting a record for the worst in 43 years. On the surface, the reasons are clear—escalation of US-Iran conflict, blockage of the Strait of Hormuz, soaring oil prices, rising inflation expectations, the Federal Reserve declaring the end of rate-cut cycles, and the dollar index breaking above 100. But the real story is actually more complex.

Recently, I’ve been thinking, why was this plunge so unusually fast? The key lies in CME changing its margin system on January 13. It shifted from a fixed amount to a percentage of contract value, raising gold margin from 8% to 9%, and silver from 15% to 18%. The change seems minor, but in a rapidly rising market, a percentage-based system means that as prices go higher, the required margin increases, creating a self-reinforcing deleveraging mechanism. On January 31, over 67 million ounces of silver paper contracts were forcibly liquidated within minutes, marking a rare single-day volatility in nearly 50 years.

Leveraged trading liquidations were the core driver of this round of decline. Gold had risen from $2,600 to over $5,000 within 12 months, with a large amount of leveraged capital piled on the long side. When the conflict escalated, pushing oil prices higher rather than gold, these crowded long positions became the first to be sold off. The technical breach of the 50-day moving average triggered a cascade of stop-loss orders.

This script is actually not unfamiliar. It has played out at least three times in the past 46 years. In 1980, Volcker pushed the federal funds rate to 20%, causing gold to fall from $711 to $304, a 57.2% decline over 456 trading days. In 2011, QE tapering combined with a strengthening dollar caused gold to drop from $1,999 to $1,049, a 44.6% decline over more than four years. The 2020 correction was milder, with vaccine rollout and rebound in US Treasury yields causing gold to fall 18.6%, but it regained its upward momentum after seven months. This time, the 2026 correction has already fallen about 27%, surpassing 2020 and approaching 2011 levels. But in terms of speed, it’s the fastest of the four—completing a 21% retracement in just five trading days.

Silver’s performance has been largely overlooked. In April 2025, the gold-silver ratio once broke above 100:1, with silver lagging severely. Then silver began a surge, with an annual increase of 147%, and the gold-silver ratio compressed to 46:1—its lowest level since 2013. But on January 31, everything reversed: silver plummeted 13.7% in a single day, with a cumulative decline reaching 41.1%, far exceeding gold’s 26.6% decline in the same period. Silver’s nine-month rally was almost entirely wiped out in less than two months.

Silver’s dual nature determines this asymmetry. On one hand, photovoltaic panel capacity grew 18% year-over-year, so industrial demand for silver remains structurally expanding. On the other hand, leverage in the silver market is much higher than in gold, and the margin increase from 15% to 18% directly triggered large-scale forced liquidations. The fundamental industrial outlook hasn’t changed, but financial leverage has already collapsed.

Changes in capital structure are also worth noting. From 2022 to 2023, global gold ETFs experienced continuous net outflows, but central banks kept buying gold, adding over 1,000 tons. In 2024, ETF flows turned positive but remained small, while central banks continued high-volume purchases. In 2025, the situation changed dramatically: global gold ETF net inflows reached 801 tons, a record, doubling assets under management to $559 billion. But retail investor funds started to flee. On March 4, SPDR Gold Trust saw a single-day net outflow of $2.91 billion, the largest since 2016, with 25 tons of gold withdrawn in just seven days.

Meanwhile, in 2025, global central banks bought 863 tons of gold, down 21% from the previous year but still well above the average from 2010 to 2021. Poland’s central bank, with a purchase of 102 tons, remained the world’s largest buyer for the second consecutive year, followed by Kazakhstan and Brazil.

The current question is, for oil-importing central banks, prioritizing responses to oil price shocks may outweigh continuing to accumulate gold. The IEA’s March oil market report states that the blockade of the Strait of Hormuz has reduced global crude oil supply by about 8 million barrels per day, nearly 8% of global demand. Coupled with tariffs and other factors pushing inflation expectations higher, member countries have initiated the largest strategic reserve releases in history—400 million barrels. In the past three peaks, global central banks were not net buyers of gold. In 2025, they bought 863 tons, but the 2026 figure remains unknown. JPMorgan maintains its year-end 2026 gold price target of $6,300, but the actual trend will depend on geopolitical developments and the next moves of central bank policies.