Gold has fallen by more than a quarter from its January peak, while silver has dropped by more than half. On the surface, this looks like nothing more than a natural cooling of an overheated rally. But beneath the price decline, the structural imbalance in the physical market has not only failed to heal 鈥?it has continued to worsen. Central banks set a record for gold purchases in a quarter when prices fell 14%, and silver recorded a supply deficit for a sixth consecutive year with the shortfall widening further. The gap between paper prices and physical supply-demand fundamentals is becoming increasingly difficult to ignore.
Gold touched an all-time peak of $5,608 per ounce (spot price) in January 2026 before reversing lower, falling to around $4,112 by mid-October 鈥?a decline of approximately 27% from the peak. Silver briefly touched $121.62 on January 29 before being halved, and currently trades near $56. Meanwhile, according to World Gold Council data, global central bank gold purchases in the second quarter of 2026 reached 288.9 tonnes, the highest on record for that period and a surge of 62% year-over-year 鈥?precisely during a quarter when gold prices fell 14%, running entirely counter to the behavior pattern of momentum-chasing funds.
The physical silver market is sending similarly anomalous signals. According to the Silver Institute's "World Silver Survey 2026," the global silver market is expected to record a supply deficit of 46.3 million ounces in 2026, further widening from 40.3 million ounces in 2025, marking a sixth consecutive annual deficit, with mine supply showing virtually no growth. Prices plunged 53%, yet the deficit grew rather than shrank 鈥?a combination that far exceeds what an ordinary commodity correction could explain.
Paper markets quelled the run, but the physical market never healed
To understand what this decline actually changed, one must return to what happened in the London market in January of this year.
Silver prices surged 25% within two weeks, and the gold-to-silver ratio fell below 45 鈥?a historically rare extreme level. Silver touching its all-time high of $121.62 did not come from a gradual repricing but from a classic short squeeze: physical silver lease rates spiked, the forward curve inverted, and the "exchange-for-physical spread" between the London OTC market and the New York futures market widened dramatically 鈥?meaning the two markets temporarily assigned vastly different prices to the same metal.
The squeeze then collapsed as expected. High prices simultaneously accomplished three things: suppressing marginal demand, attracting new short sellers into the market, and theoretically encouraging holders to release inventory. However, according to analysis, in the 2026 squeeze, the first two mechanisms operated weakly 鈥?solar manufacturers and connector makers would not halt production simply because raw material costs rose 30%, and most private holders in Asia remained on the sidelines. What truly pushed prices down was "paper selling" in the futures and options markets. The physical metal did not become more abundant; two months later, the full-year deficit forecast was revised even higher.
This logic determines the nature of the pullback: the decline flushed out speculative positions, not the supply-demand gap itself. The London market's physical float inventory was never replenished.
Central banks: the hand accelerating purchases during the decline
Central bank gold accumulation is nothing new. What truly deserves attention is the composition of buyers since 2022, their behavioral patterns, and their impact on market structure.
The World Gold Council's ninth central bank gold reserves survey in 2026 showed that 89% of central bank officials expect global gold reserves to continue rising over the next 12 months, and 45% of surveyed institutions said they would expand their own holdings 鈥?the latter hitting a record high. A contemporaneous HSBC survey of reserve managers showed that approximately 40% of respondents planned to increase holdings, with most explicitly stating that current price levels would not deter them. Among a community of reserve managers typically known for buying on dips, this orientation of "continuously adding regardless of price" is one of the most notable structural shifts in recent years.
Morgan Stanley research shows that gold's share of global central bank reserves has surpassed that of U.S. Treasuries for the first time since 1996. Poland, Turkey, India, Singapore, and several Gulf and Central Asian institutions have all increased holdings. This trend is not ideologically driven but more closely resembles proactive defensive repositioning by reserve managers after witnessing the freezing of Russian reserves.
One impact is often underestimated: official reserve gold almost never flows back into the market. By some estimates, the London physical market's available float inventory is estimated at between 3,000 and 4,000 tonnes, and a single quarter's net absorption of 289 tonnes is equivalent to withdrawing roughly 8% of tradeable stock, and that metal will not return to market prices. It should be noted that the above float inventory figures are estimates, not publicly disclosed data, and should be treated as order-of-magnitude references rather than precise measurements.
Of course, the data itself carries considerable uncertainty. The World Gold Council significantly revised down its first-quarter 2026 central bank gold purchase figure from an initial estimate of 244 tonnes to 57 tonnes 鈥?a revision of nearly 80%. Single-quarter data has limited reliability, but a directional trend spanning four years and cross-verified by multiple sources will not be overturned by one data revision 鈥?the institutions that said they would continue increasing holdings have not yet stopped.
Silver: a metal being "consumed to exhaustion" by industry
Silver's predicament is mechanistically different from gold's, and therefore more deeply significant.
Once mined, gold remains almost permanently in the world 鈥?bars, coins, jewelry, official vaults, ETFs 鈥?and the market's "deficit" is essentially a redistribution among holders rather than an absolute shortage, as the metal can always be found at some price. Silver is different. Of annual industrial consumption, 50% to 60% goes into finished manufactured goods 鈥?solar panels, high-speed connectors, automotive sensors, server motherboards 鈥?with silver measured in grams or even milligrams permanently dispersed, typically uneconomical to recover at end of life. This means the silver market is every year irreversibly converting a portion of monetary metal into industrial products, and over the years, the monetary stock available for squeezing grows ever smaller 鈥?this is precisely the structural backdrop for January's brief price break above $121 and the gold-to-silver ratio collapsing below 45.
Silver's largest single industrial end market is photovoltaics, which consumed 186.6 million ounces of silver in 2025. Affected by solar cell technology iteration (thinner silver paste lines, alternative metallization routes), photovoltaic silver usage fell 6% in 2025 and is expected to decline another 19% to approximately 151 million ounces in 2026, the largest annual drop on record. By conventional logic, a 19% contraction in the largest demand segment should narrow the market deficit 鈥?yet the opposite occurred, with the 2026 deficit forecast widening further.
Filling the gap left by photovoltaics is the rapid expansion of data center construction, high-voltage transmission equipment, electric vehicles and charging infrastructure, and grid upgrades. The common feature of these applications is that silver's functionality is irreplaceable. Photovoltaics can be redesigned to reduce silver usage, but high-voltage switch contacts and silver-sintered power modules in server racks have virtually no substitution space 鈥?because silver's conductivity here is a functional requirement, not a decorative attribute. The market reality in 2026 is this: the 35 million ounces of demand lost from photovoltaics has been replaced by demand that is less substitutable and more diversified across applications, and the deficit has risen rather than fallen.
Supply side: a market that takes a decade to respond
Prices should summon new supply. In the silver market, this mechanism has almost failed.
Ore grades in major producing regions have been declining for decades, and the energy and capital investment required to extract the same output has increased significantly. Environmental approval cycles now exceed ten years in multiple jurisdictions. Financing costs are far higher than 2010s levels, raising project thresholds and eliminating marginal mines. Even so, mining companies, after two rounds of capital discipline, may hold cash but remain unwilling to expand production.
The demand-side expansion cycle is vastly different: a data center goes from financing to operation in 18 months, a solar plant can be grid-connected within a year, and an EV production line can be retrofitted within two years. The industrial consumption expansion cycle is roughly one-fifth that of the supply-side expansion cycle.
Mining executives' caution has its own internal logic: prices in 2026 surged then halved within 18 months 鈥?the worst possible price signal, one that encourages overinvestment at the top and then punishes it. First Majestic Silver CEO Keith Neumeyer has publicly stated multiple times that silver should trade between $100 and $200 per ounce, and repeatedly warned in interviews with Kitco News: "The market will wake up one day and find there is no silver left to buy."
Trade policy: a trigger not yet pulled
Geopolitics and trade policy bring additional fragility to the physical market.
In November 2025, silver and platinum group metals were added to the U.S. Geological Survey's critical minerals list and triggered a national security investigation under Section 232 of the Trade Expansion Act of 1962. The market immediately began pricing in silver import tariffs 鈥?the direct consequence being that metal would be absorbed into U.S. warehouses, London's available float inventory would contract further, the exchange-for-physical spread would widen again, replicating the 2021 London metal drain that nearly endangered the futures delivery mechanism.
On January 14, 2026, the U.S. President issued a proclamation deciding not to impose tariffs on refined critical minerals and their derivatives for now, and directed relevant agencies to advance trade negotiations. Silver had already been excluded from the reciprocal tariff list.
In other words, the world's largest economy has already established a policy mechanism capable of redirecting global silver flows at any time, and it has not yet been activated. Once triggered, it would replay the 2021 liquidity shock in a market with even smaller float inventory and six consecutive years of deficits 鈥?with more fragile initial conditions.
Fiscal debt: the most enduring macro foundation for precious metals
All of this sits atop an increasingly strained fiscal backdrop.
According to estimates by the U.S. Committee for a Responsible Federal Budget, the U.S. federal deficit for fiscal year 2026 will reach approximately $2 trillion, with interest expenses of about $1.1 trillion, or 3.4% of GDP 鈥?a record high, making it the second-largest expenditure item in the federal budget, exceeding defense and Medicare spending. The Congressional Budget Office projects that net interest expenses as a share of GDP will climb from 3.3% to 4.6% by 2036.
The link between precious metals and fiscal deficits is not the simple logic of "deficits cause gold prices to rise" 鈥?short-term correlations between the two are unstable. A more persuasive proposition is narrower: a government facing rising interest costs and slowing growth has extremely limited policy tools 鈥?raising taxes harms investment, cutting spending may trigger recession, and borrowing only defers the pressure. The option that ultimately remains is often the gradual erosion of currency purchasing power. The asset that hedges this risk is precisely physical metal with no counterparty.
Ray Dalio's 2020 Davos remark that "cash is trash" is no longer merely a provocative aphorism six years later, as interest expenses as a share of GDP continue to rise.
Three outcomes
Combining all these factors, this situation has three evolutionary paths:
Orderly repricing. Gold and silver grind higher over several years, with occasional pullbacks similar to 2026; mining investment eventually responds, new supply arrives in the early 2030s, deficits gradually narrow, and precious metals stabilize at a higher price plateau. This is the smoothest scenario and, based on existing evidence, the most probable 鈥?it requires no new catalyst, only continued central bank accumulation and continued industrial demand growth 鈥?and both conditions are currently on track.
Disorderly squeeze. Tightening tariff policy, policy missteps, or sudden inventory depletion trigger another squeeze similar to January's 鈥?but this time the market has smaller float inventory and thinner market-maker balance sheets. Prices overshoot violently, industrial consumers are forced to buy regardless of price, and the shock transmits down the supply chain to manufacturing costs, component prices, and consumer end products. This scenario requires nothing new 鈥?only a repeat of what has already happened once.
Systemic rupture. A broader monetary or fiscal event 鈥?a failed Treasury auction, a major emerging economy currency crisis, disorderly sovereign debt repricing 鈥?forces a step-change in the reserve system. Gold rises several-fold, official accumulation becomes something akin to a rush, and the current reserve system architecture is formally rather than informally abandoned. This is the least probable but most far-reaching scenario, and the moment when the term "precious metals crisis" truly earns the weight of its name.
Supporting this overall judgment are three independent pillars: sovereign accumulation, physical deficits, and fiscal deterioration. All three must collapse simultaneously for the thesis to be falsified 鈥?and current signs indicate that none is significantly wavering.
Comments