On September 10, silver closed at approximately $64.58 an ounce, while gold closed near $4,392. The resulting gold-silver ratio was roughly 68: it took about sixty-eight ounces of silver to buy one ounce of gold.
Two days earlier, the arithmetic looked different. Using the September 8 reference prices of $4,390 gold and $66.20 silver, the ratio was about 66.3. 1
That difference is not cosmetic. It is the argument.
Citigroup’s published-or-compiled bullish case places silver near $110, while J.P. Morgan’s current published forecast places silver at $63 in the fourth quarter of 2026. 2 One camp sees the ratio falling toward 45–55 as industrial tightness joins the monetary bid. The other sees it remaining above 70 as gold retains the safer reserve-asset premium.
Silver is not a smaller gold. It is gold’s more argumentative cousin: part monetary metal, part industrial input, and entirely capable of making both sides of a trade look foolish.
The ratio is the spine
The gold-silver ratio is not a law of nature. It is a price relationship shaped by monetary demand, industrial demand, interest rates, inventories, and investor positioning.
In the modern era, the ratio has often occupied a broad range around 60 to 70, although it has periodically moved far outside that band. J.P. Morgan’s market analysis notes that the ratio fell below 45 in late January 2026 before returning toward 70 as physical tightness eased and interest-rate expectations became more hostile to silver. 2
The January episode was extraordinary. Silver rose above $121 an ounce on January 29, 2026, then suffered a peak-to-trough decline of roughly 38% as leveraged positions were liquidated. 1 The metal did not merely experience a correction. It demonstrated the peculiar physics of a smaller market whose physical supply is slow to respond and whose paper positioning can reverse in an afternoon.
The ratio provides a cleaner way to read the forecasts:
- A ratio near 45–55 implies silver materially outperforms gold.
- A ratio above 70 implies gold remains the stronger monetary asset.
- A ratio near 66–68 says the market has not yet chosen decisively.
If gold remained at $4,392 and the ratio fell from 68 to 50, silver would rise to approximately $87.84: an increase of about 36%. If the starting ratio is 66, the same move produces a gain of roughly 32%. That is the mechanical force behind bullish silver forecasts. They do not necessarily require gold to reach $6,000. They require silver to close part of the historical gap.
What the Street is really debating
The following table comes from Regatta’s internal research notes compiled in September 2026. It includes desks that published views on both metals, so each implied ratio is the firm’s own pairing rather than a mixture of unrelated forecasts.
| Firm | Gold forecast | Silver forecast | Implied G/S | Gold vs. Sept. 8 spot | Silver vs. Sept. 8 spot |
|---|---|---|---|---|---|
| Citigroup | $5,000, 6–12 months | $110, H2 | 45.5 | +13.9% | +66.2% |
| Goldman Sachs | $4,900, year-end | $92.50 midpoint of $85–$100 | 53.0 | +11.6% | +39.7% |
| Bank of America | $4,360, 2026 average | $85.93, 2026 average | 50.7 | −0.7% | +29.8% |
| HSBC | $4,750, year-end | $75, 2026 average | 63.3 | +8.2% | +13.3% |
| UBS | $5,200, 12 months | $70, December 2026 | 74.3 | +18.5% | +5.7% |
| Commerzbank | $5,000, year-end | $67, year-end | 74.6 | +13.9% | +1.2% |
| J.P. Morgan | $4,500, Q4 | $63, Q4 | 71.4 | +2.5% | −4.8% |
| RBC | $5,203, Q4 bull case | $77.48, 2026 average | 67.2 | +18.5% | +17.0% |
The table should be treated as a research summary, not as a single homogeneous public forecast set. J.P. Morgan’s $63 Q4 forecast and approximately $70 annual average are publicly documented. 2 Several other precise figures above: including the Goldman midpoint, UBS pairing, and some horizon descriptions: could not be independently matched to dated primary reports during preparation and remain compiled from Regatta’s internal research notes. They require compliance verification before publication.
The scenarios not included in the table deserve the same caution. Bank of America’s $8,000 gold and Citigroup’s $110–$150 silver are scenarios rather than base-case entries. UBS’s $80 silver reference is a September 2027 projection, not a year-end 2026 forecast.
The one-line division is straightforward:
- Silver-outperformance camp: Citigroup, Goldman Sachs, Bank of America.
- Gold-over-silver camp: UBS, Commerzbank, J.P. Morgan.
- Middle: HSBC and RBC.
The question is not whether silver has a story. It is which story is in charge.
Why silver can outperform gold
Silver has two engines
Industrial and technology uses accounted for approximately 657.4 million ounces of silver demand in 2025, against total demand of about 1.13 billion ounces: roughly 58% by calculation. 1 Gold’s technology demand is a much smaller share of total demand, generally below 10% in World Gold Council classifications. 3
Silver therefore receives the monetary bid when investors seek a hedge against currency debasement, debt, or geopolitical risk. But it also participates in demand for photovoltaics, electronics, electric vehicles, grid infrastructure, semiconductors, data centers, and other electrical applications. The combination can be powerful.
Gold has one dominant economic personality. Silver has at least two, and they do not always agree.
The official-sector gold bid is the structural monetary foundation. Central-bank gold buying has remained an important feature of the market since 2022, while central banks do not accumulate silver at comparable scale. 3 BRICS is better understood as the political face of this reserve diversification than as the daily mechanical driver of every silver price move.
The dollar-devaluation thesis also needs precision. It is not necessarily a forecast of a dramatic DXY collapse. The more defensible formulation is “gold up in every fiat.” Gold can rise against the dollar, euro, yen, rupee, and other currencies even when the DXY is stable or temporarily stronger. When discussing “dollar strength,” this article means the DXY, not gold’s performance against other currencies.
The deficit survived the solar haircut
The World Silver Survey 2026, produced by Metals Focus for the Silver Institute, reports a 40.3-million-ounce deficit in 2025 and forecasts a 46.3-million-ounce deficit in 2026: the fifth and sixth consecutive annual shortfalls, respectively. 1
The cumulative deficit from 2021 through 2026 is estimated at approximately 762.1 million ounces. 1
The important detail is not merely that the market remains short. It is that the market remains short after the largest growth segment has already begun to thrift.
Photovoltaic silver demand reached approximately 197.5 million ounces in 2024, fell to 186.6 million ounces in 2025, and is forecast at 151 million ounces in 2026: a decline of about 19% in the latest forecast. 1
That is the honest wrinkle in the bullish case. Solar manufacturers are using less silver per cell, testing copper substitutes, and changing production designs. Yet the market is still projected to run a 46.3-million-ounce deficit after the cut.
If the largest growth buyer can reduce its consumption so sharply and the market remains short, the rest of the demand structure may be tighter than the solar headline suggests.
Supply does not answer quickly
Global mine production rose to 846.6 million ounces in 2025 and is forecast to ease slightly to 844.1 million ounces in 2026. 1
But most silver is not produced because a miner woke up and decided to produce more silver. In 2025, primary silver mines accounted for only about 26% of global mine supply. The remaining 74% came largely as a byproduct of lead-zinc, copper, and gold mining. 1
That structure makes silver supply inelastic on the way up and sticky on the way down. A higher silver price does not instantly create a new copper mine. It may improve the economics of a primary silver project, but permitting, financing, construction, and commissioning remain long processes.
The frequently repeated claim that China controls 60–70% of global silver refining could not be verified against the Silver Institute or USGS material reviewed for this draft and is not used as a factual premise here. The documented point is more modest but still important: China produced approximately 112.8 million ounces of mined silver in 2025, and Chinese bullion exports reached a record 162 million ounces that year. 1
Physical tightness can outrun paper positioning
The Silver Institute reported that total physically backed silver exchange-traded-product holdings rose to approximately 1.3 billion ounces at the end of 2025, while non-ETP London inventories fell sharply during the 2025 liquidity squeeze. Metals Focus estimated that only about 136 million ounces of London inventory remained freely available outside ETP allocations at the end of September 2025. 1
Regatta’s late-August 2026 internal inventory work placed SLV holdings near 494 million ounces. That figure is internal analysis, not a substitute for the official ETP series, which reported iShares Silver Trust holdings of 529 million ounces at year-end 2025. 1
The distinction matters. Metal held inside an exchange-traded product or shipped into Asian physical channels is not necessarily available for the next Western bid. It may exist, but it is not the same thing as an immediately mobilizable float.
The documented market structure is price discovery through venues such as COMEX and the LBMA, combined with periods in which paper-market positioning lags physical tightness. The World Silver Survey also documents a collapse in CME open interest from 768 million ounces at year-end 2025 to 577 million ounces by February 2026, alongside falling CME inventories. 1
That does not prove a permanent bank-set ceiling. It does not prove that BRICS “broke COMEX” in one dramatic event. It does show that futures positioning, open interest, exchange inventories, lease rates, and physical availability can move out of alignment.
The inventory figures cited here refer to total reported CME inventories, not registered-only, eligible-only, or a claim about deliverable stock in every category.
Why silver often fails to keep outperforming
Silver’s dual demand is also its central weakness.
No official-sector floor
Central banks buy gold as a reserve asset. They do not buy silver by the hundreds of tonnes as a normal part of reserve management. The official-sector bid therefore tends to stabilize gold more directly than silver during monetary stress. 3
Real yields and the Fed
Silver is the high-beta twin of gold. When the DXY rises and real yields increase, non-yielding assets become less attractive. Silver, with its smaller market and industrial exposure, often absorbs the sharper decline.
J.P. Morgan’s current analysis explicitly links its more cautious silver forecast to a potential Fed hiking environment, weaker solar demand, and the normalization of the gold-silver ratio toward 70 in the second half of 2026 and 75 in 2027. 2
Demand destruction is real
High prices are not merely a reward. They are also a negotiation with the buyer.
The Silver Institute expects industrial demand to decline another 3% in 2026, largely because PV demand is falling. Jewelry and silverware demand are forecast to decline by 16% and 20%, respectively, as elevated prices force lighter products and deferred purchases. 1
J.P. Morgan has separately estimated that solar silver demand could fall by roughly 30% in 2026, or about 60 million ounces, as thrifting becomes more widespread. 2
Substitution is no longer theoretical
Major solar manufacturers are adopting lower-silver pastes, silver-coated copper, zero-busbar designs, and copper electroplating. The World Silver Survey expects average silver loadings to continue falling through 2027, even though high-reliability applications may retain silver’s advantages. 1
That creates an asymmetry:
- Substitution can damage the phase-three investment story for silver miners.
- It does not eliminate the monetary role of an ounce held in a physically backed vehicle.
- It can reduce industrial demand even while investment demand remains strong.
Volatility cuts both ways
The same market that produced a price above $121 also produced a peak-to-trough decline of approximately 38% in a short period. 1
Silver does not offer a free lunch. It offers a smaller table with sharper elbows.
The next six to twelve months
| If this happens | Likely ratio effect |
|---|---|
| Fed hike or higher real yields | Ratio widens; gold holds up better |
| Fed pause and renewed ETF inflows | Ratio tightens; silver catches up |
| PV thrifting exceeds the current 19% decline forecast | Ratio widens |
| AI, grid, electronics, and automotive demand offset PV losses | Ratio tightens |
| Total COMEX inventories and London free float keep shrinking | Ratio may tighten sharply |
| Gold rises toward $4,800–$5,000 without an industrial pickup | Ratio widens first, then may tighten later |
The key variables are not mysterious. Watch the DXY, real yields, gold’s official-sector demand, ETP flows, physical premiums, lease rates, and the direction of solar loadings.
Chinese physical tightness also deserves priority over Western narrative. The East Asian market is not merely responding to a Western paper signal; it is a major fabrication and investment center. In Regatta’s reading, Chinese physical conditions often lead the Western paper market, although the claim should be treated as an analytical framework rather than a claim that every daily move originates in China.
Silver miners are not smaller gold miners
A silver miner is not simply a gold miner with a more exciting ticker.
The first difference is supply. In 2025, non-primary silver production reached approximately 625.5 million ounces, while primary silver mines accounted for roughly 26% of mine output. 1
The second difference is the listed universe. Regatta’s internal screen found roughly fifteen silver-focused names above $500 million in market capitalization after applying a primary-revenue filter. That is internal analysis, not an official industry census. The broader universe includes gold miners, base-metal producers, streamers, royalty companies, and developers that happen to produce silver.
The third difference is cost accounting. Metals Focus reported a 2025 global primary-silver AISC of $12.21 per ounce on a byproduct-accounting basis, with margins of approximately $27.81 per ounce. 1
Selected company screens and market commentary often show a broader practical cost range around $16–$25 per ounce, depending on whether the analysis uses company-reported AISC, silver-equivalent ounces, byproduct credits, sustaining capital, and corporate costs. That range should not be confused with the Metals Focus global average.
For Regatta’s metals analysis, cost pressure is assessed through the “original PPI”: year-over-year prices for intermediate processed and unprocessed goods: rather than relying solely on the final-demand headline. The practical kill-switch is older and simpler: is an ounce buying more diesel, labor, and steel than it did twelve months ago? That is fundamentally a pre-1980 PPI question.
SIL illustrates the problem. As of September 8, 2026, Wheaton Precious Metals represented approximately 23.8% of the Global X Silver Miners ETF, followed by Pan American Silver, Coeur Mining, and Hecla. 4 SIL is therefore not a pure operating-miner basket. It contains a substantial streaming-company component, and streaming can dampen operating leverage while reducing some mine-level risk.
Regatta’s internal performance work also found that SILJ had beaten SIL by approximately 17 percentage points during the most recent twelve-month rally window, while trailing the metal over longer periods. Those figures are internal analysis and require verification before publication. The broader point is less fragile: junior miners are more exposed to financing conditions, development risk, dilution, and sentiment than an ounce of silver.
The four phases
| Phase | Silver market | Senior miners | Juniors |
|---|---|---|---|
| Stealth | Deficit builds while the ratio remains wide | Trade like high-cost gold or base-metal producers | Financing is difficult; they often lag the metal |
| Recognition | Silver breaks old ceilings and deficits remain visible | Margin expansion begins to appear | First financings return, but execution risk remains |
| Participation | Industrial tightness combines with investment flows | SIL and senior producers can outperform silver | SILJ and developers can move several times faster |
| Blow-off or washout | Leverage, premiums, and paper positioning reverse | Equity leverage gives back part of the gain | Juniors become first in and first out; single-session losses can be severe |
The historical pattern
The Hunt Brothers episode of 1979–80 was a paper-market squeeze without the multi-year industrial deficit now visible in the Silver Institute’s data. It ended as a corner unwound.
The 2008–11 cycle combined monetary demand with China’s industrial expansion. Silver reached approximately $49.82 in 2011 before margin increases and a stronger dollar helped break the advance. The episode left a long shadow over silver equities, particularly juniors.
The 2020 cycle was a liquidity event. Silver miners rose with the broader reflation trade, but real yields and a stronger dollar later exposed how little durable industrial tightness had been priced into the equities.
The 2023–26 cycle is different in one important respect: it combines a multi-year deficit with a genuine blow-off episode. Regatta’s internal analysis estimates that SIL rose approximately 166% in 2025 and SILJ approximately 184%, while year-to-date 2026 performance through the analysis date showed SIL ahead of SILJ by roughly 19% versus 15%. These return figures are internal research and require final market-data verification.
That sequence: surge, air pocket, then relative senior strength: looks less like a fresh stealth phase and more like late phase two or an early failed phase three.
Three silver-specific kill switches
1. The byproduct trap
If copper and zinc prices fall sharply, much of the world’s silver supply may continue because the mines are being operated for other metals. At the same time, silver-miner earnings can deteriorate because byproduct credits shrink.
This is why copper should not be placed on the same de-dollarization chart as gold and silver. Copper’s price path is governed primarily by industrial demand, Chinese grid and electric-vehicle development, inventories, treatment charges, and mine supply: not by the monetary reserve thesis.
2. PV substitution
A second major decline in solar silver loadings could break the phase-three narrative for SILJ and single-name developers. It would not eliminate the monetary bid for physical silver, but it would reduce the industrial-tightness premium that junior equities need in order to justify ambitious valuations.
3. Streamer weight
A large streaming-company allocation makes SIL less operationally sensitive than a pure primary-miner portfolio. That may be useful for risk control, but it also means the ETF is not a direct expression of mine-level operating leverage.
The sleeve framework
The following is a framework, not individualized advice.
- PSLV: the silver-ounce component inside a broader 10–16% metals band, using Regatta’s default 60/40 to 70/30 PHYS/PSLV relationship.
- SIL and selected senior producers: an optional satellite of approximately 1–3% of an account only when SIL has beaten silver for a quarter, AISC remains well below spot, and copper and zinc are not rolling over.
- SILJ and single-name juniors: qualified-only, higher-risk exposure after SIL has already confirmed participation. It is cut first in a phase-four tape and is never a substitute for physical silver exposure.
At the September 10 close, the relevant picture was silver around $64–66, a gold-silver ratio near 66–68, a sixth projected deficit year, and a January blow-off already spent. The ounce remains the cleaner instrument. The miners remain a leveraged expression of a market that must get several things right at once.
The honest answer is concentrated in one sentence: the market is still forecast to show a 46.3-million-ounce deficit after photovoltaic demand has already fallen approximately 19%.
That is the whole ballgame.
If the deficit still binds after solar thrift, silver can outperform gold without requiring another dramatic gold repricing. If substitution, high real yields, weak industrial demand, or investment outflows overwhelm the deficit, gold remains the cleaner monetary hedge and silver becomes a volatile trading overlay.
Prudence does not require choosing one story forever. It requires knowing which story is currently paying the bill.
Sources
- Silver Institute and Metals Focus, World Silver Survey 2026
- J.P. Morgan Global Research, “Silver Prices”
- World Gold Council, Gold Demand Trends
- Global X, SIL Silver Miners ETF: holdings, performance, and fund information
- Silver Institute, “Elevated Lease Rates, Regional Liquidity Tightness, and Robust Investor Interest Resulted in Record Silver Prices in 2025”
- Silver Institute, “Silver Industrial Demand Reached a Record 680.5 Moz in 2024”
- U.S. Geological Survey, Silver Statistics and Information
- Regatta Financial internal research notes, September 2026
Disclosure
Regatta Financial LLC provides fee-only investment advisory services with no commissions on investment transactions. This article is for educational and informational purposes only and does not constitute individualized investment, tax, legal, or accounting advice; a recommendation or solicitation to buy or sell any security, fund, commodity, or other investment product; or a guarantee of future results. Precious metals, mining companies, exchange-traded products, and related investments involve substantial risk, including volatility, liquidity risk, operational risk, geopolitical risk, currency risk, and possible loss of principal. Past performance is not indicative of future results. Readers should consider their own objectives, time horizon, financial circumstances, and risk tolerance and consult appropriate professionals before making investment decisions.
Confidence statement
Confidence: High for the Silver Institute/Metals Focus supply, demand, deficit, photovoltaic, mine-production, inventory, and cost figures cited from the 2026 survey; high for J.P. Morgan’s published silver forecasts and Global X’s SIL holdings; moderate for the broader historical and market-structure interpretation; and limited for the individual bank forecast pairings and return figures identified as compiled from Regatta’s internal research notes.
Leave a Reply