Gold and Silver — Monthly Report September 2026

1. Executive summary

The precious metals complex enters September caught between two forces that normally do not operate at the same time: a central bank actively preparing to raise interest rates, and a geopolitical shock that is simultaneously stoking inflation and eroding confidence in dollar assets.

Gold is roughly 21% below the record $5,595–5,597/oz set on 29 January 2026, having bottomed briefly under $4,000 on 24 June and then recovered through a ~10% August rally. Silver, at ~$66.6, is down roughly 45% from its own January record near $121.6 but is still up more than 60% year-on-year. Both metals have re-based dramatically lower from their January blow-off, yet both remain enormously higher than a year ago. That combination — violent drawdown inside an intact secular uptrend — is the defining feature of this market.

Our central view for the next three to six months:

• Gold: constructive but range-bound near term, with medium-term upside. We see a base case of $4,200–$4,800 into year-end, with the FOMC on 16 September the single most important near-term determinant. A hike is not the bear case people assume; a hike delivered into 3.4% headline inflation and a $100 oil shock is a policy-error hedge, and gold has historically traded the credibility of the response more than the level of the rate.
• Silver: higher beta, wider distribution, lower conviction. Base case $55–$80, with genuinely fatter tails in both directions than gold. The bull case rests on physical tightness; the bear case rests on solar demand destruction. Both are live.
• The structural bid is unbroken. Central banks bought a record-for-a-second-quarter 289 tonnes in Q2 2026 — while prices were falling. The ECB confirmed in June that gold has overtaken US Treasuries as the world’s largest reserve asset. This is the floor under the market, and it is not a price-sensitive floor.
• The cyclical headwind is real. Rising nominal and real yields, a Fed that has removed all 2026 cuts from the table, and elevated opportunity cost have driven North American ETF holders out of the trade. That flow reversal is why gold is at $4,400 and not $5,500.

The single most important thing to understand about this market right now: the US 10-year yield is at ~4.8%, its highest since October 2023, and the 30-year has touched 5.34% — yet the dollar index sits near 99, close to a three-month low. Rising yields with a falling dollar is not a growth story. It is a risk-premium story, and risk premium on US sovereign credit is the most gold-positive macro configuration that exists.

2. Where we stand: the 2026 price journey

Gold

Gold posted its first monthly gain since February in July, then added roughly 10% in August. It has drifted lower in the first week of September as rate-hike odds firmed.

Silver

Silver’s 2026 has been extraordinary even by silver standards: above $113 in early January, down to $77 by February (a 32% move in weeks), a run back above $90, then a collapse into the $56–58 range in June, and a grind back to the mid-$60s.

The ratio

The gold/silver ratio compressed from roughly 80:1 to about 50:1 at silver’s January peak, then normalised back toward the high 60s. Over the last twelve weeks alone it has travelled 61.7 → 70.4 → 66.3 — a round trip far wider than its typical range, and a clean illustration of silver’s higher beta to gold in both directions.

Read-through: at ~66, the ratio is below its recent-decade norms (it hit 100:1 in April 2025) but above the long-run 40–60 band. Silver is neither obviously cheap nor obviously expensive against gold. Anyone using the ratio as a timing tool right now is working with a very noisy signal.

3. The macro regime: two engines pulling opposite ways

3.1 The Fed — the September 16 decision

Fed funds sit at 3.50–3.75%. The July FOMC held rates with a 9–3 vote, with Hammack, Kashkari and Logan dissenting in favour of a hike. Chair Kevin Warsh — sworn in on 22 May 2026 — used his debut Jackson Hole speech on 28 August to say underlying inflation had not “meaningfully improved” and that the Fed has “work to do.” Market-implied odds of a 25bp hike on 16 September swung from below 40% to roughly 58–66% depending on the venue and the day.

The picture is genuinely knife-edge, and it has moved on almost every data point:

• Hawkish inputs: August payrolls came in at +162,000 against a +56,000 consensus, with prior months revised up. Warsh’s explicit statement that he would be “hard pressed” to call financial conditions restrictive. Oil at $100. Producer prices expected to accelerate.
• Dovish inputs: Governor Waller has said he is open to holding if inflation data eases. Citi’s Hollenhorst has argued there was no consensus to hike in July and there will not be one in September. Three consecutive weak payroll prints preceded the August rebound.
• Political inputs: the Trump administration is applying overt pressure to stop the hike. This matters more for gold than the rate decision itself — see §3.4.

Two data points sit between now and the decision: August PPI on 10 September (headline expected ~5.3%, core ~4.6% — accelerating) and August CPI on 11 September (headline expected steady at 3.4%, core easing slightly to 2.4%). A hot CPI makes the hike near-certain. A soft one gives Warsh the cover to wait.

Markets also price roughly a 50% chance of a further hike in December.

3.2 The inflation impulse: Iran, oil and the supply shock

The US–Israel war against Iran began on 28 February 2026 and is now in its seventh month. As of 9 September, Brent has crossed $100/bbl for the first time since July, after US forces destroyed five Iranian tankers in the Persian Gulf in response to IRGC missile launches at a US warship, and after Houthi attacks on Saudi energy infrastructure. Crude is roughly 40% above the pre-war ~$70 level and about 60% higher on the year. US diesel hit a record $5.94/gallon.

Goldman’s Daan Struyven has said the probability of a scenario in which Gulf exports stagnate and Brent exceeds $120 is “definitely going up.” Bank of America’s Blanch has flagged $150 in a broader-infrastructure-damage scenario.

Why this matters for metals, and why the sign is not what intuition suggests: a classic geopolitical shock is unambiguously gold-positive. This one is not, because it transmits through the oil price into headline inflation, which forces the Fed to tighten, which raises the opportunity cost of holding a non-yielding asset. Through much of 2026 the metals have fallen on Iran escalation headlines. That is unusual, and it is the mechanism to understand: energy-driven inflation is currently a net negative for gold via the policy channel, unless and until it breaks growth or breaks the Fed’s credibility.

3.3 Rates, real yields and the opportunity-cost channel

• 10-year Treasury: ~4.78–4.81%, highest since October 2023
• 30-year Treasury: ~5.25%, having touched 5.337%
• 2-year Treasury: highest since November 2024

Elevated real yields have been the proximate cause of Western ETF liquidation all year. The World Gold Council explicitly attributed the weakest North American first half since 2013 to hawkish Warsh signals plus Iran-driven inflation fears pushing real yields and the dollar up together.

3.4 The dollar and the fiscal-risk premium — the part the consensus is under-weighting

Here is the anomaly. The dollar index sits near 99, having bottomed at 98.55 on 22 August, its weakest since mid-May, inside a 52-week range of roughly 95.6–101.8. It fell to a four-year low earlier in the year. Meanwhile long yields are at multi-year highs.

Under normal conditions higher US yields pull foreign capital in and lift the dollar. That relationship has broken. Investors appear to be pricing rising yields as compensation for fiscal deficits, inflation persistence and political interference in monetary policy rather than as a return on healthy growth. CNBC reported on 7 September that Treasuries are losing foreign appeal in what it called a historic capital-flow reversal.

Layer on the administration’s public campaign to prevent a rate hike, and the market has a live question about central bank independence. Bank of America’s Michael Widmer has named Fed leadership uncertainty, structural fiscal deficits, and historically low investor gold allocations as three under-appreciated upside risks for gold.

This is the mechanism through which gold gets to new highs even in a hiking cycle. It requires the market to conclude that the Fed is either behind the curve or not free to act. Both are plausible from here.

4. Gold: demand architecture

4.1 Official sector — the floor, and it is holding

This is the strongest part of the bull case and the most under-appreciated by price-focused commentary.

The H1 number is depressed by heavy Q1 selling from Turkey, Russia and Azerbaijan, not by weak buying. In Q2, Turkish disposals slowed materially and Russia was the only sizeable seller.

Buyers of note:

• Poland — largest reported buyer, +82t in H1 to 632t, still working toward a stated 700t target
• China (PBoC) — approximately 60t added in the first seven months, the strongest opening seven months since 2023; +15t in June, +20t in July; reported holdings ~2,346t at mid-year
• Uzbekistan (+16t Q2), Kazakhstan (+15t), Jordan (+6t), Czech Republic (+6t)

The critical detail: Q2 was gold’s steepest quarterly price decline in a decade, and official-sector buying accelerated into it. Reserve managers are not momentum traders. Goldman’s own note assumes a central bank floor of roughly 50 tonnes a month. UBS still models 750–1,000t of annual official demand.

And the structural marker: the ECB’s June 2026 International Role of the Euro report confirmed that gold has surpassed US Treasuries as the world’s largest reserve asset. That is a milestone with a long half-life.

4.2 Western investment — the swing factor, and the reason for the drawdown

Global gold ETF holdings peaked at a record 4,176t on 27 February 2026. As of the last confirmed monthly data (July), holdings stood at 4,068t with AUM of $530bn, after a +$3bn July inflow that broke a two-month outflow streak. Year-to-date through July, global inflows were +$11bn / +39t — with Asia the largest contributor, Europe second, and North America still in net outflow.

The regional split is the whole story:

• North America: H1 outflows of $7.7bn, the weakest first half since 2013. June alone saw -$5.5bn. Roughly 298t of ETF gold was estimated to be underwater at the June lows.
• Europe: the swing buyer, led by UK and German funds on fiscal and safe-haven concerns.
• Asia: strongest H1 on record earlier in the year, though with intermittent outflows.

Q2 2026 saw net ETF redemptions of 45t globally.

Forward view: the August price rally almost certainly pulled ETF flows positive again; the WGC’s August report should confirm this. But the more important question is whether North America returns. Those investors bought on a rate-cut thesis that has been comprehensively invalidated. They will not return on the same thesis. They return on either (a) a growth scare, or (b) a Fed-credibility scare. Watch which.

4.3 Consumer demand — price-rationed, not destroyed

Q2 2026 global jewellery demand fell to 278t, down 5% q/q and 17% y/y. Indian jewellery demand dropped 15% y/y to 75.1t. Consumers are adapting through lighter-weight pieces, old-for-new exchange, and rotation into lower-premium investment products rather than exiting gold.

Bar and coin investment was 307t in Q2, holding up year-on-year.

Total Q2 demand including OTC was 1,269t, flat y/y. H1 demand was 2,522t, +2% y/y, at a record H1 value of $380bn.

The signal: demand composition has shifted from adornment to allocation. That is what a monetary-demand regime looks like, and it is more price-resilient than a jewellery-led one.

4.4 Supply

Mine production grew 2.4% y/y in Q1 2026 per WGC data, but the GDX top-25 producers saw output fall 10.7% y/y — a reminder that headline supply growth and listed-producer supply growth are different things. Mine production growth remains structurally pinned in the 1–2% range regardless of price.

4.5 Producers — the leveraged expression

Miners have delivered the sharpest round trip of any gold exposure this year. GDX was -17.8% YTD on 20 July, then rallied hard through August to close above $102 on 21 August. On 4 September, miners fell roughly four times harder than bullion on the strong payrolls print.

Underlying economics are exceptional. In Q1 2026 Agnico Eagle realised $4,861/oz against all-in sustaining costs of $1,483/oz — a cash margin near $3,400 an ounce. Newmont posted record free cash flow of $3.1bn. Both ended the quarter with roughly $3bn in net cash. Sector valuations compressed to decade lows even as earnings hit records.

Read-through: miners are a high-beta expression of the gold view, not a substitute for it. They amplify both directions and add operational and jurisdictional risk. In a $4,400 gold environment they are printing money; in a $3,800 environment the margin compression is non-linear.

5. Silver: two markets in one metal

Silver’s difficulty is that roughly 59% of its demand is industrial. It gets repriced by monetary factors and by manufacturing cycles, and in 2026 those two channels are pulling in opposite directions.

5.1 The balance

The Silver Institute’s World Silver Survey 2026 (researched by Metals Focus) sets the picture:

2026 marks the sixth consecutive annual deficit. Cumulative drawdown from above-ground stocks since 2021: 762.1 Moz.

A note on a number circulating in the market: some outlets have cited a 2026 deficit of 215 Moz. The figure in the World Silver Survey 2026 and in Metals Focus/Reuters reporting is 46.3 Moz (earlier preliminary estimate: 67 Moz). We use the survey figure. Anyone building a thesis on the larger number should verify its provenance.

5.2 Demand composition — where the bear case lives

• Industrial fabrication: forecast down 2–3% to roughly 640–650 Moz, a four-year low.
• Solar PV — the swing variable. Global installations continue to rise, but ongoing thrifting (using less silver per cell) and outright substitution mean PV silver demand is falling. This is the single most important negative development of 2026.
• Data centres / AI infrastructure: a genuine and growing offset. Electrification and EV manufacturing likewise.
• Jewellery: -9% to 178 Moz, lowest since 2020, on Indian price rationing.
• Silverware: -17%.
• Physical investment: +18–20% to 227 Moz, a three-year high, with US retail demand expected to rebound roughly 57% after three years of decline.

The 2026 pattern is investment demand mopping up what industry and jewellery leave behind. That is a less stable equilibrium than industrial-led demand, because investment demand is reflexive — it grows when prices rise and evaporates when they fall.

5.3 The physical market and the squeeze question

October 2025 remains the reference event. London silver lease rates spiked to roughly 39% (normal: below 1%) as the deliverable free float collapsed. By end-September 2025, physically backed products accounted for 83% of London inventories, leaving only 17% genuinely available for market operations — a free float estimated near 136 Moz against average daily OTC turnover of roughly 450 Moz. That squeeze drove the January 2026 record.

Where we are now: metal flowed back from New York, lease rates normalised to the low single digits by mid-Q1 2026 and lower since, and the non-ETP share of London holdings recovered to roughly 24%. February 2026 LBMA data showed 27,065 tonnes (~870 Moz), down from 894.4 Moz at end-2025. The Survey described conditions as feeling similar to historical norms.

The LBMA is considering moving to weekly silver inventory publication (currently monthly) precisely to give earlier warning of future tightness.

Forward view: the squeeze mechanism has not been dismantled, only relieved. The free float is structurally thinner than pre-2021, the deficit continues, and any renewed ETP inflow surge re-tightens the float. A repeat is not the base case, but it is a live tail risk with an enormous payoff distribution. Watch lease rates and the non-ETP share of London stocks as the early-warning pair.

5.4 Analyst capitulation

August 2026 saw the sharpest downward revision cycle silver has seen in years:

J.P. Morgan expects the gold/silver ratio to normalise toward 70 in H2 2026 and around 75 in 2027, on the view that physical tightness unwinds and higher rates raise the opportunity cost of holding non-yielding assets.

Note the important detail: silver is currently trading above J.P. Morgan’s Q4 forecast. The bank cut to a level the market has already rejected. Treat the August revision cycle as a marker of sentiment capitulation rather than a reliable price path.

6. Forward view: scenarios to year-end 2026 and into 2027

Gold

12-month view: we expect gold higher, in the $4,800–$5,400 region by end-2027, on the assumption that fiscal deterioration and reserve-diversification demand persist regardless of the near-term rate path. This is broadly where the sell-side clusters (see §7).

Silver

Note the asymmetry in the distribution. Silver’s bull tail is fatter and further out than gold’s because the float is thin; its bear tail is also more probable because half its demand is cyclical. Position sizing should reflect that this is a wider distribution, not merely a shifted one.

7. Where the sell-side sits

Forecasts have been revised repeatedly and in both directions this year. Dates matter enormously.

Gold

Reuters’ poll of 31 analysts put the 2026 median near $4,916.

The pattern to notice: the mid-year revision wave was overwhelmingly downward (Goldman, J.P. Morgan, HSBC, Citi in June–July). The late-summer wave has been upward (UBS, Citi, Commerzbank, Wells Fargo). The sell-side turned bearish into the June low and bullish into the August rally. This is a lagging indicator, and it should be used as one.

Silver

Post-August-revision range is roughly $55 (bear) to $100 (bull) for parts of 2026, with the cluster around $63–$74. Pre-revision bull cases (Goldman $85–100, Citi $110 H2, Bank of America’s $135–309 ratio-compression scenario) are now well outside consensus and should be treated as scenario analysis rather than forecasts.

8. Catalyst calendar

9. Indicator dashboard — what to actually watch

For gold, ranked by signal value:

1. 10-year TIPS real yield. The cleanest single driver of Western positioning. Falling real yields would mark the turn.
2. DXY versus long-end yields. If yields rise and the dollar falls together, that is the fiscal-risk-premium regime and it is gold-positive. If they rise together, it is the opportunity-cost regime and it is gold-negative. This one relationship tells you which market you are in.
3. North American ETF flows. The marginal Western buyer. Currently absent.
4. Monthly PBoC and NBP reserve reports. The official-sector floor, reported with a lag but reliably.
5. COMEX managed-money net positioning. Positioning was already stretched into the September payroll print.

For silver, additionally: 6. London silver lease rates. Normal is <1%. Anything sustained above 5% signals the float is tightening. 7. Non-ETP share of LBMA London holdings. 17% triggered the October 2025 squeeze; ~24% currently. This is the squeeze gauge. 8. Solar PV silver loadings per cell. The structural demand question. Thrifting data is slow-moving but decisive. 9. The gold/silver ratio itself — as a confirmation tool, not a timing tool.

10. Risks to our view

Risks to the constructive gold case:

• A genuine Iran de-escalation would knock $20–30 off oil, pull headline inflation down fast, and remove both the inflation hedge bid and the geopolitical bid at once. This is the most underpriced bear risk.
• A Fed that hikes decisively and is believed re-anchors expectations and lifts real yields for the right reasons.
• Central bank buying is reported with a lag and is not guaranteed to persist at 289t/quarter. H1 2026 was already the weakest first half since 2022 on a net basis.
• Positioning is not washed out. There is still ETF gold underwater from the January peak that can be sold.

Risks to the silver case specifically:

• Thrifting and substitution in solar are permanent, not cyclical. Once a manufacturer re-engineers a cell to use less silver, that demand does not come back at any price.
• The deficit is small relative to above-ground stocks. 46 Moz against a market that holds hundreds of millions of ounces in London alone is not, by itself, a forcing constraint. It matters because the available float is thin — which is a different and more fragile argument.
• Investment demand doing the heavy lifting means the demand base is reflexive and can reverse violently.

Risks to the bear case (i.e. what would force us more bullish):

• Warsh does not hike, and the market reads it as political capture rather than data dependence.
• Foreign official demand for Treasuries deteriorates further; the long end backs up while the dollar falls.
• Hormuz closes.

11. Positioning considerations

This section describes how the analysis maps onto portfolio construction. It is not investment advice, and I am not a financial adviser — your own circumstances, tax position, time horizon and risk tolerance should govern any decision.

Framing that follows from the analysis above:

• Gold and silver are not one trade. Gold is currently a monetary and sovereign-credit hedge. Silver is that plus a levered bet on industrial demand. Treating them as a single “precious metals” allocation obscures the fact that they can and do diverge sharply — as the 12-week ratio round trip demonstrated.
• The structural and cyclical horizons disagree, and both are right. The central bank bid argues for a multi-year floor. The rate path argues for near-term range-trading. An allocation sized for the structural view and an allocation sized for the tactical view are different sizes.
• Silver requires smaller sizing for the same risk budget. Its realised volatility this year — $113 to $77 to $90 to $56 to $66 — is not gold’s volatility. A position sized as though it were will produce a risk exposure several times larger than intended.
• The event risk around 16 September is binary and known. Anyone with a view on the FOMC outcome should recognise that the market has already priced roughly 60% of one, meaning the asymmetry favours the no-hike surprise for metals.
• Miners amplify, they do not diversify. Record margins at $4,400 gold are real, and valuations are at decade lows on earnings. But the 4:1 downside beta on the September payrolls print is the same coin.

12. Bottom line

Gold at $4,395 is not a broken bull market. It is a bull market that ran 45% in four months into January, priced in a Fed easing cycle that never arrived, and has spent seven months unwinding that specific error while the structural bid underneath it kept buying.

The near-term path runs through the 11 September CPI print and the 16 September FOMC. Our base case is a hike, a range-bound autumn, and a resumption higher as the market re-focuses on the thing that actually matters: a US fiscal and monetary configuration in which long yields rise while the dollar falls and foreign capital steps back. Gold has already passed US Treasuries as the world’s largest reserve asset. Reserve managers made that decision while prices were falling. That is the tell.

Silver is the same trade with a manufacturing cycle bolted on and a physical market thin enough to break. Higher expected return, materially wider distribution, and a solar demand question that nobody has satisfactorily answered.

Sources

Price and macro data: TradingEconomics (gold, silver, US 10-year, DXY, 8–9 Sep 2026); JM Bullion; Yahoo Finance / Forbes Advisor daily precious metals coverage; Federal Reserve H.15.

Fed policy: CNBC (28 Aug, 31 Aug, 5 Sep 2026); Bloomberg (28 Aug); Morningstar; PBS NewsHour; Fortune; Axios (3 Sep); Charles Schwab FOMC commentary (July 2026); Marketplace; CME FedWatch and Kalshi as reported.

Geopolitics and energy: NBC News (8–9 Sep 2026); CNN Business (9 Sep); CBS News; Forbes (9 Sep); CNBC commodities (9 Sep).

Gold fundamentals: World Gold Council — Gold Demand Trends Q1 and Q2 2026, Central Bank Gold Reserves Survey 2026, monthly Gold ETF Flows (May, June, July 2026 editions); State Street Monthly Gold Monitor (August 2026); WisdomTree Gold Monthly (August 2026); ECB International Role of the Euro (June 2026, as reported).

Silver fundamentals: The Silver Institute / Metals Focus — World Silver Survey 2026 and February 2026 preliminary forecast; Reuters (10 Feb 2026); Kitco (15 Apr 2026); LBMA London vault data through February 2026.

Analyst forecasts: J.P. Morgan Global Research (silver, 13 Aug 2026); Goldman Sachs (19 Jun 2026); UBS (23 Jul, Aug 2026); Citi (24 Aug 2026); Morgan Stanley (20 Aug 2026); Bank of America; ING; Commerzbank; HSBC; Wells Fargo Investment Institute; Reuters analyst poll — as compiled and reported by Reuters, Bloomberg, Yahoo Finance and specialist aggregators.

Producers: VanEck GDX fund data; Benzinga (18 May 2026); Zeal LLC Gold Miners’ Q1’26 Fundamentals (15 May 2026); company reporting for Newmont and Agnico Eagle.

This report is for informational and educational purposes only. It is not investment, financial, legal or tax advice, and it is not a recommendation to buy or sell any security or commodity. Forecasts are estimates, not guarantees. Precious metals are volatile and you can lose money. Some figures — particularly forward-looking supply/demand balances and analyst targets — are revised frequently and may be stale by the time you read this. Verify current prices and data before acting, and consult a qualified adviser about your own situation.

Terry S. - Editor in Chief

Terry brings over 25 years of experience in stock and options trading, having actively navigated markets since 1999. A seasoned trader who has weathered multiple market cycles—from the dot-com boom and bust through the 2008 financial crisis to today’s dynamic markets—he combines deep market knowledge with technical expertise.

As a developer and digital creator, Terry has built and launched multiple financial websites and trading tools, bridging the gap between complex market analysis and accessible financial information. His unique perspective comes from hands-on experience on both sides of the screen: as an active trader executing strategies and as a developer creating platforms that serve the trading community.

Terry’s coverage focuses on actionable market analysis, options strategies, and technical insights drawn from real-world trading experience. He specializes in identifying market trends, analyzing options flows, and translating complex market movements into clear, practical insights for traders at all levels.

When not analyzing markets or developing new tools, Terry continues to actively trade and test strategies, ensuring their analysis remains grounded in current market realities.

Terry’s Verified LinkedIn Profile | About StockMarketWatch