Gold-Silver Ratio Widens Over the Week: Apparent Flight to Safety Masks Internal Reallocation Within Precious Metals
From September 11 to 14, the gold-silver ratio widened from 67.31 to 68.15. Silver fell significantly harder than gold; this was not driven by panic liquidation, but rather by silver’s industrial attributes being priced separately during a week of rising interest rate expectations.
—— ◆ ——
| Core Summary · September 11–14 (Friday to Monday) |
|---|
| The gold-silver ratio widened from 67.31 to 68.15. Silver’s decline was markedly larger than gold’s, triggered by hotter-than-expected U.S. CPI data released on September 11, which pushed the probability of a September rate hike to approximately 90%.<br><br>Over the past 12 weeks, the gold-silver ratio has oscillated between 61.7 and 70.4 for two full cycles. This current widening is merely the latest segment of this significant volatility and does not yet constitute a trend-breaking breakout.<br><br>Gold’s support primarily stems from central bank purchases, which are insensitive to interest rates (244 metric tons in Q1 2026, up 17% quarter-over-quarter). In contrast, approximately 59% of silver demand comes from industrial uses, making it more sensitive to interest rate cycles.<br><br>Minli Square Assessment: This widening appears to be a short-term phenomenon resulting from the concentrated pricing of rate hike expectations, rather than a systematic exit of capital from safe-haven assets. The supply deficit in silver has persisted for six consecutive years and remains unchanged.<br><br>The key watershed level is 70.4, the high point over the past 12 weeks. Only if this previous high is truly broken should the medium-term logic for silver be reassessed. |
Catalyst I: September 11–14 – Three Trading Days, Gold-Silver Ratio Widens by 0.84
On Friday, September 11, the U.S. August CPI data was released, showing a month-on-month increase of 0.4% and a year-on-year increase of 3.4%. Core CPI rose 0.3% month-on-month, slightly above market expectations. Consequently, the CME “Fed Watch” tool immediately raised the probability of a September rate hike to approximately 90%. Prior to this data release, markets had largely expected the Federal Reserve to hold rates steady at its upcoming meeting. Upon the data’s publication, position adjustments on trading desks were nearly instantaneous. On the same day, spot gold stood at $4,385.61 per ounce, essentially flat; spot silver stood at $64.90 per ounce, dropping sharply by 2.58%. Silver, with its stronger industrial attributes, clearly struggled to withstand the intensifying impact of rising rate hike expectations.
By Monday, September 14, this divergence continued to widen: according to FXStreet data, the gold-silver ratio further widened from Friday’s 67.31 to 68.15, with silver’s decline once again outpacing gold’s. Coupled with sudden geopolitical news from the Middle East early that morning, market sentiment remained cautious. Capital exhibited distinct “picky” behavior within safe-haven assets—buying gold while shunning silver.
Notably, this divergence occurred against a backdrop where the overall atmosphere for precious metals was not particularly poor. The Federal Reserve’s policy meeting is scheduled for September 15–16. Institutions such as Everbright Futures described the week as “Super Fed Week” in their commentary on September 14, suggesting that gold volatility may intensify further. Domestic financial media also reported early on the 14th that sudden geopolitical developments regarding Iran briefly drove down U.S. stocks, gold, and silver simultaneously. In other words, this was not a one-sided rally in gold accompanied by the abandonment of silver, but rather two asset classes experiencing different slopes of decline under the shock of the same macroeconomic event.
| Date | Spot Gold | Spot Silver | Gold-Silver Ratio |
|---|---|---|---|
| September 1 (Tuesday, Reference) | $4,374.54/oz, -1.60% | $64.76/oz, -2.69% | 67.55 (Widened from previous day’s 66.84) |
| September 11 (Friday) | $4,385.61/oz, +0.06% | $64.90/oz, -2.58% | 67.31 |
| September 14 (Monday) | Price stabilized, weak | Decline exceeded gold’s | 68.15 (Widened by 0.84 from Friday) |
Background II: Not the First Time – Over the Past 12 Weeks, the Ratio Has Oscillated Twice
Extending the timeline reveals that this current widening is nothing new. Over the past 12 weeks, the gold-silver ratio has swung significantly between 61.7 and 70.4, almost covering the entire range twice—hitting a low of 61.7 in mid-June, surging to 70.4 in early August, falling back to 66.3 on September 8, and reaching the current reading of 68.15. This is simply the latest data point in this period of significant volatility, not an abrupt emergence of a new trend.
Viewed within a longer historical coordinate system, this number is not extreme: the 50-year average of the gold-silver ratio is approximately 60; the annual average for 2025 reached a high of 85.7, compressing significantly only after silver prices broke through the $50 threshold late in the year; in early 2026, when silver prices surged near $121, the ratio compressed to lows in the 40s. Compared to these extremes, the current level around 68 remains in the middle of the fluctuation range for the past year.
This severe volatility itself is a direct reflection of the silver market’s structure: the spot and futures markets for silver are far smaller than those for gold. Therefore, capital inflows or outflows of similar magnitude exert a much larger impact on price movements. This explains why the gold-silver ratio rarely exhibits smooth trends like gold prices themselves, instead frequently displaying large-scale oscillations. Understanding this prevents one from simply interpreting a weekly widening or narrowing as confirmation of a trend reversal.
| Timeframe | Ratio | Context |
|---|---|---|
| 50-Year Average | ~60 | Long-term equilibrium level calculated by the Silver Institute |
| 2025 Full Year Average | 85.7 | High at the start of the year, compressed significantly at year-end as silver broke $50 |
| Early 2026 (Silver near $121) | Compressed to 40s | Silver’s gains initially far exceeded gold’s |
| Mid-June 2026 | 61.7 (12-week low) | Phased low point |
| Early August 2026 | 70.4 (12-week high) | Surged again within 5 weeks |
| September 8, 2026 | 66.3 | Fell back within a week |
| September 14, 2026 | 68.15 | Latest reading discussed in this article |
Breakdown III: Who Is Actually Buying Gold and Silver?
Behind the surface-level widening of the gold-silver ratio lies the fact that the two metals are priced by entirely different groups of capital. The core buyers of gold are national central banks—global central banks net purchased 244 metric tons of gold in Q1 2026, a 17% quarter-over-quarter increase. This buying pressure is driven by political and reserve security considerations and is largely insensitive to interest rate levels, objectively setting a price floor for gold. Conversely, pricing power for silver rests more heavily with the industrial sector: according to calculations by the Silver Institute, industrial applications accounted for approximately 59% of global silver demand in 2026, covering scenarios such as photovoltaics, electric vehicles, and data centers. This portion of demand is more sensitive to economic cycles and interest rates.
This divergence is also reflected in the “nature of source” of the capital: central bank gold purchases are often completed via over-the-counter agreements, unaffected by open market sentiment, with daily purchase volumes remaining relatively stable. In the silver market, procurement from manufacturing sectors such as photovoltaics and electronics adjusts flexibly based on order cycles and raw material inventory strategies. Once prices experience sharp fluctuations, industrial purchasers often choose to wait or delay procurement, further amplifying the volatility of silver prices during shifts in macroeconomic expectations.
| Dimension | Gold | Silver |
|---|---|---|
| Core Buyers | Central banks (net purchases of 244 tons in Q1 2026, +17% quarter-on-quarter) | Industrial sector (approximately 59% of demand comes from photovoltaics, electric vehicles, data centers, etc.) |
| Sensitivity to Interest Rates | Low—central bank gold purchases are not driven by interest rates, establishing a price floor | High—industrial demand follows the economic cycle and is more sensitive to rate hikes |
| 2026 ETF/Investment Fund Flows | Record outflows in March, reversing to $6.6 billion in inflows in April | Photovoltaic silver demand fell 19% to 151 million ounces; investment silver (coins/bars) expected to rise 18% to 257.6 million ounces |
| 2026 Supply Side Performance | Q1 mine production reached 885 tons, a record high, +2% year-on-year | The sixth consecutive year of supply deficit, with an estimated shortfall of 46.3 million ounces |
Lesson Four: "Silver has greater elasticity than gold" only holds true during rate-cutting cycles
Many investors are accustomed to viewing silver as a "leveraged version of gold"—rising more sharply when prices go up. This perception is generally valid during periods of rising rate-cut expectations, but the logic completely reverses when rate-hike expectations intensify: silver’s industrial attribute shifts from a driver to a drag, and capital prioritizes reducing silver’s pro-cyclical risk exposure rather than gold’s. This explains why silver’s decline was dozens of times deeper than gold’s following the release of CPI data on September 11.
Applying this logic back to the 2025 trend clarifies the picture: throughout 2025, as rate-cut expectations fluctuated upward, silver prices surged over 100% at one point within the year, compressing the gold-silver ratio continuously from its highs at the start of the year until silver itself approached $121 per ounce, exhausting short-term gains before beginning a phased correction. In other words, silver’s "high elasticity" label essentially stems from its dual dividend of enjoying both safe-haven premiums and industrial demand premiums in a loose monetary environment; once macroeconomic expectations shift toward tightening, this dual dividend becomes dual pressure, meaning it rises fast but falls just as quickly.
| Phase | Direction of Gold-Silver Ratio | Core Logic |
|---|---|---|
| Period of Rising Rate-Cut Expectations | Narrowing (silver’s gains usually exceed gold’s) | Silver’s industrial attribute + relatively low price make it more favored by capital under loose expectations |
| Period of Rising Rate-Hike Expectations (e.g., this round in September) | Widening (silver’s declines exceed gold’s) | The industrial attribute becomes a burden instead, with capital prioritizing the reduction of silver’s pro-cyclical exposure |
| Extreme Uncertainty Phase (e.g., early March 2020 pandemic onset) | Extreme widening (spiking to 125 at one point) | Industrial demand collapsed faster than safe-haven buying could establish itself, causing silver to fall faster than gold |
Comparison Five: Another contrast on the supply side: Gold is increasing production, while silver remains in deficit
If one looks only at the demand side, it is easy to form the impression that "silver is scarcer," but incorporating the supply side reveals a sharper contrast. Gold’s Q1 mine production hit a record 885 tons, a 2% year-on-year increase; however, silver’s mine production has remained stable at approximately 830 million ounces annually for many years, showing virtually no growth. The market has been in a state of supply deficit for six consecutive years, with cumulative deficits since 2021 reaching approximately 762.1 million ounces—an amount equivalent to consuming an entire year’s worth of mine production from above-ground inventories.
This contrast is significant because it indicates that the widening of the gold-silver ratio is not due to "silver’s fundamentals deteriorating," but rather reflects a change in "short-term capital preference." If silver were truly experiencing oversupply and shrinking demand, weakening prices should have been accompanied by inventory accumulation. However, the latest calculations from the Silver Institute show that 2026 will still be the sixth consecutive year of supply deficit, with the scale of the gap expanding compared to the previous year. This stands in stark contrast to the phenomenon of "silver prices weakening in the short term," representing two largely unrelated threads. Conflating the two easily leads to overly pessimistic conclusions such as "silver is done for."
| Dimension | Gold | Silver |
|---|---|---|
| 2026 Q1 Mine Production | 885 tons, a record high, +2% year-on-year | Global mine production has remained stable at approximately 830 million ounces/year for many years, with near-zero growth |
| Supply-Demand Status | Overall tight balance maintained, with central bank gold purchases providing continuous support | Sixth consecutive year of supply deficit; 2026 estimated deficit in the range of 46.3 million ounces |
| New Supply Elasticity | Major gold producers generally expect full-year production declines and rising mining costs | New mine development requires 5–10 years; short-term deficits cannot be easily filled by new supply |
Assessment Six: Mineral Cube’s Judgment: More likely a pricing phenomenon of the rate-hike week, not a signal of capital withdrawal
Mineral Cube’s judgment is that this round of gold-silver ratio widening is more likely a short-term phenomenon where the market concentratedly priced in rate-hike expectations ahead of the September FOMC meeting, rather than a signal of systematic capital exiting silver or the precious metals sector. The reasoning is that silver’s six-year supply deficit has not been falsified by any new data, and the long-term demand logic for industrial applications in new scenarios such as photovoltaics and data centers remains unchanged—what is under short-term pressure is price elasticity, not fundamentals.
Viewed from another angle, the gold-silver ratio acts more like a "capital preference detector" rather than a "master switch for risk aversion sentiment"—it reflects how the same pool of safe-haven capital is allocated between gold and silver, rather than whether that capital is flowing into or out of the precious metals sector as a whole. During the week of September, central bank gold purchases did not stop, and the Silver Institute’s annual supply-demand report did not downgrade forecasts; the only change was in short-term interest rate expectations. This supports the judgment of "internal reallocation" rather than "collective exit."
| Observation Angle | Evidence Supporting "Short-Term Noise" | "Structural Signals" Requiring Vigilance |
|---|---|---|
| Price Action | Touched interval extremes twice within 12 weeks before reverting, showing clear mean-reversion characteristics | If it subsequently breaks above the 12-week high of 70.4, it may confirm a new upward range |
| Supply Deficit | Silver’s sixth consecutive year of deficit has not been falsified; the long-term logic remains intact | If photovoltaic substitution accelerates further, industrial demand may continue to weaken |
| Central Bank Behavior | The pace of gold purchases continues, showing no sign of a broader retreat in safe-haven demand | If monthly central bank gold purchase data weakens, the stability of gold’s price floor needs to be re-evaluated |
Outlook Seven: 3 Tracking Signals
Rather than obsessing over weekly ratio fluctuations, it is more valuable to establish several signals that can be continuously verified—single-day price movements are never sufficient to define a trend.
| Tracking Signal | Frequency | Significance |
|---|---|---|
| Whether the gold-silver ratio breaks through 70.4 | Daily | Determines whether this widening is range-bound oscillation or a trend-breaking breakout |
| Silver industrial demand breakdown (photovoltaics, NEVs, data centers) | Quarterly (Silver Institute/Metals Focus reports) | Assesses whether industrial demand is continuing to weaken or is only temporarily pressured |
| Monthly central bank gold purchase data | Monthly (World Gold Council) | Assesses whether gold’s price floor remains solid |
Tracking Point 1: Treat 70.4 as the key watershed; a breakout above previous highs is required to reassess the medium-term logic for silver. Tracking Point 2: Do not focus solely on the gold-silver ratio figure; disaggregated data on silver’s industrial demand reflects trend changes earlier than the ratio itself. Tracking Point 3: Only when central bank gold purchase data shows consecutive months of sequential weakness constitutes a genuine signal requiring vigilance regarding potential loosening at the bottom of gold prices; single-month fluctuations do not constitute a basis for judgment. Tracking Point 4: Place every significant fluctuation in the gold-silver ratio within the context of comparison against the 61.7–70.4 range observed over the past 12 weeks, avoiding misreading range-bound oscillations as trend reversals.
—— ◆ ——
The widening of the gold-silver ratio can never be summarized merely by the phrase “retreat of safe-haven flows”; it acts more like a detector, revealing shifts in capital preferences within precious metals rather than the overall presence or absence of capital.
— END —