Gold Holds $4,374 as Flat PPI Cuts September Hike Odds to 40% and Silver Tests $67

Gold traded at $4,374.03 on Thursday, August 13, after December futures opened at $4,468.80 per troy ounce — flat against Wednesday's settlement — and drifted to $4,441.10 by 7:53 a.m. Eastern. Spot quotes ranged between $4,374 and $4,410 through the European session, with the metal easing into the 8:30 a.m. producer price release.

Wednesday delivered the strongest session in two months. Spot gold closed at $4,424.44, up $54.24 or 1.24%, its highest level in roughly ten weeks. The advance began before the U.S. inflation print landed, with London hours already showing bullion 1.1% higher near $4,414.63 as traders pared rate-hike expectations and leaned on the metal to hedge renewed Middle East supply risk. That detail matters: gold reached a two-and-a-half-month high while market-implied odds of a September Federal Reserve hike still sat near 48%. Metal advanced without a decisive dovish trigger.

This is the fourth consecutive session with an opening print above $4,400. Gold is up nearly 10% month over month and 30.40% year over year, yet only 1.92% higher year to date — an arithmetic gap that captures the entire 2026 story in two numbers. The 52-week range runs from $3,311.46 to $5,595.46.

The week's reference points are tight. Gold entered the August 10 week at $4,343.07. It closed Wednesday at $4,424.44. Futures settled near $4,468. Spot pulled back to the $4,374 to $4,390 zone Thursday morning. That is a $130 band containing five sessions of trade, with each pullback finding buyers roughly $50 below the prior high.

Silver has been the sharper mover and the better tell. September futures opened Thursday at $65.46, down 0.4% from Wednesday's $65.70 close, and traded at $65.23 by 8:11 a.m. Eastern. Spot silver reached $67.06 on Wednesday, up 2.84% and its best level since June, before slipping to $64.48 on Thursday — a 1.30% decline from $65.32 — after touching a seven-week high earlier in the session. Silver is up 70.42% year over year and 10.43% over the past month.

The gold/silver ratio compressed to roughly 66 on Wednesday's move, its tightest reading in months, which historically marks the point where industrial demand rather than monetary demand is setting the marginal price in the complex.

PPI at 0.0% Delivered the Headline Gold Wanted and the Core Print It Did Not

The July Producer Price Index for final demand was unchanged at 0.0% in the Bureau of Labor Statistics release, against a consensus calling for a 0.2% increase. The annual rate fell to 4.7% from 5.5% in June, undershooting the 4.9% expected. June was revised to a 0.1% decline from the 0.3% drop originally reported.

The headline was built entirely on energy. Final demand goods fell 0.7% after a 1.4% drop in June, with final demand energy down 3.1%. Gasoline fell 5.7% and accounted for more than half the goods decline. Diesel, jet fuel, residual fuels and thermoplastic resins all moved lower. At the unprocessed intermediate stage, crude petroleum collapsed 11.9% and unprocessed energy materials fell 7.4%. Final demand foods fell 0.9%.

The component that governs Fed policy went the other direction. Final demand less foods, energy and trade services — the series feeding core PCE — accelerated to 0.4% in July from 0.1% in June, and holds a 4.7% annual rate. Final demand services rose 0.2%, but services less trade, transportation and warehousing advanced 0.6%. Portfolio management prices jumped 6.5%. Final demand construction rose 2.2%. Stage 4 intermediate demand, closest to the consumer, rose 0.6% and sits 6.7% above year-ago levels, while services for intermediate demand rose 0.5% with a 5.1% annual rate.

Wednesday's Consumer Price Index set the same pattern. Headline CPI rose 0.1% with the annual rate easing to 3.4% from 3.5%, and core CPI printed 0.2% monthly and 2.5% annually — the slowest core reading since March 2021. Gasoline fell 2.9%.

For gold, this is a genuinely awkward combination. Two consecutive cool headline prints have trimmed September hike odds from near 50% to roughly 40%, which is the mechanical driver behind the move from $4,343 to $4,424. But the disinflation came from energy, and gold does not rally durably on falling oil. The metal's structural bid requires either accelerating inflation with a constrained central bank, or falling real yields. Cool headline inflation with hot core services delivers neither: it removes the inflation-hedge case while keeping real yields elevated.

That contradiction explains why gold gained 1.24% on the CPI print and then gave back roughly $50 into PPI rather than extending. The market is trading the Fed path, not the price level.

The 2026 Collapse: $5,595 on January 29 to $3,940 in June

Gold set its all-time high at $5,595.46 on January 29, 2026, with some feeds marking the peak at $5,602.23. That capped one of the most extraordinary runs in the metal's history — a roughly 60% gain in 2025, the best annual performance since 1979, with 53 separate all-time highs recorded during the year. Gold moved from $3,865 in October 2025 to $5,595 in January 2026, a 45% advance in approximately four months.

Then it broke. By April 13, gold had fallen roughly 15% to around $4,728. By late June it slipped below $4,000 for the first time since November 2025, touching the $3,940s. By July 21 spot sat near $4,046, approximately 28% below the January peak — the steepest quarterly correction since 2013. The Q2 2026 average LBMA price settled at $4,506, down roughly 14% over the quarter.

Three forces produced it, arriving in sequence.

The first was the Hormuz blockade. Oil moved above $100 per barrel and pushed March CPI to 3.3% year over year. Elevated inflation removed any possibility of Federal Reserve rate cuts, which was the exact thesis Western ETF buyers had used to accumulate through 2025.

The second was the Fed's hawkish turn under Chair Kevin Warsh. Rate-cut expectations did not merely get pushed back; they inverted into rate-hike expectations. September hike odds reached roughly 68% in late June, up from 29% a week earlier, and at one point futures priced a nearly 80% probability of a hike at some point in 2026. The dollar reached a 13-month high.

The third was positioning. Gold entered 2026 with extreme long exposure across ETFs and managed money after a parabolic 2025. When the first two headwinds hit, that positioning became fuel. North American ETF outflows exceeded $12.7 billion in March 2026 alone — the largest monthly redemption in at least five years — and gold fell $611 that month, the largest absolute monthly decline on record.

July marked the turn. Gold gained approximately 0.5%, its first monthly increase since February. August has extended that recovery by nearly 10%, taking spot from the $4,046 July low back above $4,400.

March's $611 Monthly Loss Was the Largest Absolute Decline on Record

The March collapse deserves separate examination because it defines the supply overhang gold still trades against.

A $611 decline in a single month exceeds every prior monthly loss in dollar terms across gold's entire history. In percentage terms it was severe rather than unprecedented, but the absolute figure matters for one reason: it was concentrated almost entirely in the paper market. COMEX managed-money positioning and ETF redemptions did the selling. Physical demand did not participate.

The mechanism is straightforward. Rising rate expectations lift real yields, which raises the opportunity cost of holding a non-yielding asset. For ETF investors benchmarked quarterly, the calculation is mechanical rather than discretionary: when the ten-year TIPS real yield reached approximately 1.94% in April, fixed income offered a positive real return against gold's zero, and redemptions followed automatically.

That is precisely what unfolded. Total gold ETF assets fell to $604 billion, with most of the decline a price effect rather than volume. Collective holdings eased 0.4% to 4,121 tonnes, barely below the record 4,176 tonnes set on February 27, 2026. Year-to-date net flows remained positive at nearly $17 billion even through the worst of the drawdown.

Read those two figures together and the picture changes. The financial holder stepped to the sidelines at the margin. It did not liquidate the stock. Holdings 1.3% below an all-time record after a 28% price collapse is not capitulation — it is a buyer who stopped adding and let mark-to-market do the damage.

An estimated 298 tonnes of ETF gold sits underwater at current prices, acquired during the November-to-January melt-up above $5,000. That inventory represents the overhead supply capping every rally. Global gold ETF holdings currently sit roughly 1.5% below where they started the year, and recent inflow patterns suggest selling pressure has eased without demand meaningfully returning.

COMEX managed-money positioning has held in neutral territory, registering only a modest $1.4 billion, or 8-tonne, gain through the spring. That absence of speculative length is constructive: there is little leveraged long exposure left to liquidate, which removes the primary mechanism behind another March-style event.

Real Yields Are the Whole Trade: 1.94% TIPS, 4.67% Nominal, 5% at the Long End

Every other variable in the gold market is secondary to the real yield, and the real yield is going the wrong way for bullion.

The federal funds target sits at 3.50% to 3.75% after five consecutive holds, most recently at the July 28–29 FOMC meeting, which passed 9–3 with three members dissenting in favor of a hike — the first three-way, same-direction dissent since September 2016. The two-year Treasury yields 4.18%. The ten-year sits at 4.66% to 4.69%. The thirty-year trades above 5%, with the previous auction clearing at 5.058% and another $25 billion sale scheduled for 1:00 p.m. Eastern Thursday.

Against 3.4% headline CPI and 2.5% core, those nominal yields produce a positive real return across the curve. The ten-year TIPS real yield near 1.94% earlier this year set the benchmark, and nothing in the current configuration has brought it down.

The bond market's behavior around the two inflation prints reinforces the problem. Treasuries held their gains through Wednesday's CPI, with the two-year down three basis points to 4.18% and the ten-year down three to 4.66% ahead of a $42 billion auction. By Thursday's pre-open the ten-year had backed up to 4.67% to 4.69%, surrendering most of the rally. Nominal yields refusing to fall on soft inflation data means real yields are rising, and rising real yields with cool headline inflation is the single most hostile configuration gold can face.

Rate pricing tells the same story from the other direction. September hike odds have fallen to roughly 40% from near 50%, with the probability of a hold at approximately 53.9%. But October odds exceed 53% and December sits at 73%. Cutting is not in the conversation. Consensus economist forecasts show no hike this year and half a point of cuts in 2027, which means positioning and forecasting have decoupled entirely.

July payrolls contracted by 23,000 against an expected gain near 85,000, and that miss is the strongest single argument for gold. A labor market rolling over eventually forces the Fed's hand regardless of energy-driven inflation. Continuing claims rose 24,000 to 1,801,000 even as initial claims held near 199,000, describing an economy where hiring has stalled without layoffs beginning.

ETF Holdings at 4,121 Tonnes: The Marginal Buyer Left and Has Not Come Back

The regime change in the gold market is a change in who sets the marginal price.

Through 2025, Western ETF buyers dominated the flow. Their accumulation accelerated as rate-cut expectations built, driving gold from $3,865 in October 2025 to $5,595 in January 2026. When the Iran conflict altered the Fed's calculus in March, those same buyers reversed. ETF flow swung from record inflows to record outflows within weeks, and price followed.

That buyer has not returned in size. Holdings at 4,121 tonnes against a February record of 4,176 tonnes show a stock that has barely moved, but the flow — the variable that moves price — has been absent for five months. The World Gold Council attributes the pause to investors sidelined by rangebound prices and renewed risk appetite elsewhere, noting technology ETFs booking their largest single-month intake since the start of 2024.

That last point is the competitive dynamic gold cannot escape. The S&P 500 sits within a tenth of a percent of a record at 7,748.50, having gained 3.6% in the first week of August. Full-year 2026 index earnings growth is projected at 30.0%. Capital that would historically rotate into precious metals during a geopolitical crisis has instead rotated into AI infrastructure, where the return is measurable and immediate rather than defensive and deferred.

The recovery from $4,046 to $4,424 has occurred without a corresponding surge in ETF creations, which is why the move has stalled at $4,400 to $4,470 rather than extending toward $4,600. Physical premiums and central bank purchases can support a floor. They cannot produce a trend. Only the paper market can do that, and the paper market is waiting for the Fed.

The threshold to watch is straightforward: sustained weekly ETF creations that lift holdings back above the February record of 4,176 tonnes would confirm the Western buyer has returned. Until then, every rally into overhead supply from the 298 tonnes acquired above $5,000 meets sellers who have been underwater for six months and want out near breakeven.

Year-to-date net flows near $17 billion remain positive, which distinguishes this correction from a genuine bear market. The stock is intact. The flow is not.

Central Banks Bought 244 Tonnes in Q1 and Are on Pace for 850 in 2026

Beneath the paper market, an entirely different buyer has not moved an inch.

Central banks and official institutions added approximately 244 tonnes during the first quarter of 2026. Full-year purchases are projected near 850 tonnes, in line with 2025's 863 tonnes and well above the 473-tonne annual average recorded between 2010 and 2021. Independent modeling puts the figure closer to 800 tonnes, which still represents a structural floor of roughly 25 tonnes of physical demand every week regardless of price.

The composition of that buying has broadened. The People's Bank of China has extended its accumulation streak to 19 or 20 consecutive months. Poland has remained among the largest single buyers. The central banks of Guatemala, Indonesia and Malaysia have entered the market — institutions absent for years or never present at all. The Bank of Korea began adding gold exposure to its reserve portfolio in the first quarter of 2026, its first gold-related step since 2013.

The forward-looking survey data is the most bullish on record. The annual Central Bank Gold Reserves Survey published June 16, 2026, covering 76 institutions, found that 89% of reserve managers expect global official gold holdings to increase over the next twelve months, with a record 45% planning to add to their own reserves — the most bullish result in the nine-year history of the report. A subsequent reading put the share expecting global holdings to rise at 95%. Reported selling remains concentrated in a small number of institutions managing domestic fiscal pressure.

The two buyer classes operate on incompatible time horizons, and that resolves the apparent contradiction between a falling price and a record bullish survey. ETF investors rebalance quarterly against benchmarks and respond to real yields within days. Central banks operate on decade-long reserve diversification mandates and respond to sanctions risk, dollar concentration and fiscal trajectory. The first group sets the price. The second sets the floor.

A more cautious read exists. At least one major commodities research team trimmed its gold forecast in June citing moderating central-bank demand alongside ETF flows. Purchases of 850 tonnes against 863 tonnes in 2025 is technically a decline, and if the pace slows toward the pre-2022 average of 473 tonnes, the structural floor weakens materially.

Gold Passed Treasuries as the World's Largest Reserve Asset at 27% of Holdings

The single most consequential development of 2026 received almost no attention because it happened while the price was collapsing.

The European Central Bank confirmed in its June 2026 International Role of the Euro report that gold has surpassed U.S. Treasuries as the world's largest reserve asset. Gold now accounts for 27% of global central bank holdings against 22% for Treasuries. Central bank Treasury holdings stand near $3.5 trillion.

That crossover occurred through two mechanisms operating simultaneously: rising gold prices lifted the mark-to-market value of existing holdings, and sovereign institutions kept buying through the entire drawdown. Neither mechanism required a single ETF creation.

The structural drivers behind it remain fully intact. U.S. fiscal deficits are running at 6% to 7% of GDP, and the July budget statement came in high. Total household debt stands at $18.8 trillion with credit card balances at $1.26 trillion. The thirty-year Treasury above 5% reflects supply pressure that has nothing to do with the policy rate. Reserve diversification away from the dollar continues, and the sanctions precedent that accelerated it after 2022 has not been reversed.

A bull market ends when the structural drivers that created it reverse. For gold those drivers are real yield suppression, fiscal expansion, central bank diversification and de-dollarization. Exactly one of the four has reversed: real yields are elevated rather than suppressed. That is a cyclical headwind operating against three structural tailwinds that remain in place.

The distinction is not academic. It determines whether $4,046 in July was a bear market low or a correction low within an intact secular advance. The 28% drawdown from January is comparable in magnitude to gold's 2013 correction, which preceded a multi-year bear market — but 2013 featured falling central bank demand, disinflation, and a Fed exiting emergency policy. The 2026 configuration inverts all three: record official-sector demand, inflation stuck at 3.4%, and a central bank debating whether to tighten further.

If ETF selling resumes and sentiment-driven redemptions continue, prices could test the lower fair-value bound near $3,895 before structural buyers absorb the supply. That figure is the practical floor implied by the physical market rather than a technical level.

Silver at $64.48 After a Seven-Week High and a Gold/Silver Ratio of 66

Silver has outperformed gold decisively and now carries a different set of drivers.

Spot silver reached $67.06 on Wednesday, up 2.84%, its best level since June. Thursday brought a pullback to $64.48, down 1.30% from $65.32, after touching a seven-week high early in the session. September futures opened at $65.46 against a $65.70 close and traded at $65.23 by 8:11 a.m. Eastern. Silver is up 70.42% year over year and 10.43% over the past month, with year-over-year growth having run as high as 173.3% on May 14.

The gold/silver ratio compressed to roughly 66 on Wednesday's advance. That compression is the market pricing industrial demand rather than monetary demand, and the underlying data supports it. Approximately 58% of silver demand is industrial. Solar-panel manufacturing and electricity-grid investment have both expanded, and Chinese imports of silver-bearing ores jumped 62.5% year over year in June to 219,000 tonnes.

Silver's market is roughly one-tenth the size of gold's, which makes it structurally more volatile in both directions. The same industrial exposure that drove the 70% year-over-year gain becomes a liability when rate-hike expectations rise and recession concerns follow, because the industrial demand component contracts precisely when the monetary component should be supporting price. Silver fell to $58.03 in late June, roughly 29% below its January high, mirroring gold's decline almost exactly despite the different demand structure.

Projections calling for silver to reach $90 as investment demand takes over rest on the assumption that the monetary bid returns while the industrial bid holds. That requires the Fed to shift from a hiking debate to an easing path without a growth contraction — a narrow window.

For gold traders, silver's relative strength is the most useful leading indicator available. A ratio compressing through 66 toward 60 confirms genuine risk appetite in the complex and typically precedes gold breaking to new range highs. A ratio expanding back above 70 signals the industrial bid failing and usually accompanies gold testing range lows. The ratio is currently the cleanest expression of whether the August recovery is real.

The Hormuz Variable: Why an Active War Has Been Bearish for Gold

The most counterintuitive fact of 2026 is that gold fell 28% during an active Middle East war, and the mechanism deserves precise explanation because it determines what happens next.

Fighting between the United States and Iran intensified from late February. U.S. and Israeli forces launched strikes on February 28. Attacks on vessels in the Red Sea and Gulf of Oman have continued, and the Strait of Hormuz has remained disrupted. The blockade sent oil above $100 per barrel at its peak and pushed March CPI to 3.3%.

Conventional analysis says war is bullish for gold. In 2026 the opposite held, because the war's primary transmission channel was inflation rather than fear. Higher oil prices meant higher headline CPI, which prevented the Federal Reserve from cutting, which kept real yields elevated, which drove ETF redemptions from the exact buyer that had powered the 2025 rally. The war worked directly against gold's most rate-sensitive demand channel while rate-insensitive central banks continued buying at a steady pace that could not offset the outflow.

The current configuration has partially inverted. July PPI energy fell 3.1%, crude petroleum dropped 11.9%, and CPI energy declined 1.5%. Brent shed 2% Thursday to $87.04 and WTI fell 2% to $81.32, with OPEC lowering its 2026 demand growth forecast and the International Energy Agency now projecting demand falling 1.6 million barrels per day this year — 510,000 barrels per day worse than its July estimate. Falling oil removes the inflation constraint, which reopens the path to Fed easing, which is bullish for gold.

The unresolved piece is the strait itself. Talks on reopening have produced no clear breakthrough. Pakistan's defense minister has indicated the U.S. and Iran are close to an arrangement, while the U.S. administration maintains it retains total control — a claim private shipping data disputes, showing traffic well below normal.

That creates a genuinely two-sided risk for gold, which is unusual. A negotiated reopening collapses oil further, accelerates disinflation, brings forward Fed easing, and lifts gold through the real-yield channel. An escalation spikes crude, revives the inflation constraint, keeps the hiking debate alive, and pressures gold through the same channel that produced the March decline. The safe-haven bid works in one direction; the rate channel works in the other, and the rate channel has been larger all year.

Technical Structure: $4,336 Support, $4,468 Futures, $4,645 as the August Ceiling

The technical picture is constructive on the medium-term chart and stretched on the short-term one.

Immediate support sits at $4,347 to $4,336, the zone the market is currently probing after Wednesday's rejection near $4,424. Below that, $4,224 marks the weekly range low, and $4,120 to $4,084 represents the projected month-end zone under a corrective scenario. The structural floor sits at $4,046, the July 21 low, with $4,000 as the psychological level that broke in late June for the first time since November 2025.

Resistance is layered tightly. The immediate ceiling runs $4,441 to $4,468, where December futures opened and where the metal has stalled four sessions running. Above that, $4,449 marks the daily range high and $4,460 the near-term target ceiling. The August upper boundary projects to $4,645, and the consolidation band for the next session runs $4,376 to $4,441.

The multi-week structure is a well-defined range that price has now reached the top of. That upper boundary coincides with the upper edge of a short-term ascending channel, creating a confluence resistance that requires genuine flow to break rather than momentum. Four consecutive opens above $4,400 without a close above $4,470 describes a market pressing against supply rather than clearing it.

Technical indicator aggregates have shifted to a strong buy rating on the medium-term timeframe, reflecting the recovery from $4,046 and the reclamation of key moving averages during the August advance. Estimated pivot sits at $4,115, with the trend rated higher.

The clean decision levels: a daily close above $4,470 opens $4,520 and then the $4,645 August ceiling. Failure at $4,441 and a break of $4,336 targets $4,224, then $4,120. The monthly projection band spans $3,580.75 to $4,645.91, which is unusually wide and reflects how much of the outcome depends on a single Fed decision on September 16.

Positioning offers an edge here that price does not. With managed money neutral and speculative length absent, a break above $4,470 has limited fuel from short covering but also limited resistance from long liquidation. The move that follows a genuine break should be slower and more durable than the parabolic advance of late 2025.

Positioning: Managed Money Neutral, 298 Tonnes Underwater, Volatility Compressed

The positioning data explains why gold has traded in a $400 band for six weeks after a $1,600 collapse.

COMEX managed-money exposure has held in neutral territory since spring, with a modest $1.4 billion, 8-tonne increase through May and no standout driver identified in attribution modeling — part of the residual move likely sitting in over-the-counter markets rather than exchange-traded positioning. That neutrality is the defining feature of the current market. The speculative excess that made January's peak possible and March's collapse inevitable has been fully purged.

The consequence cuts both ways. There is no crowded long position left to liquidate, which removes the mechanism behind another record monthly decline. There is also no crowded short position to squeeze, which removes the mechanism behind a violent upside break. What remains is a market that must be moved by actual flow rather than by positioning unwinds, and actual flow has been absent.

The 298 tonnes of ETF gold acquired above $5,000 and now underwater is the specific overhead the market trades against. Every advance toward $4,470 encounters holders who bought during the November-to-January melt-up and have spent six months down 15% to 20%. That cohort does not need a target — it needs a level close enough to breakeven to justify exiting, and each successive rally brings more of it within reach.

Implied volatility across the complex has compressed alongside the range. That reads as stabilization but functions as fragility: less protection is in place if the range resolves, and a market with compressed volatility and neutral positioning gaps rather than trends when it finally moves.

Physical demand provides the counterweight that paper positioning does not. Indian retail rates rose Thursday with 22-karat quoted at ₹14,195 to ₹14,240 per gram and 24-karat between ₹15,485 and ₹15,535, taking gold past ₹1.5 lakh and silver above ₹2.5 lakh in local terms. Physical buying at record local-currency prices during a dollar-price correction indicates the emerging-market bid remains intact and is not waiting for lower levels.

The composite read: a market with no leverage, persistent physical demand, an absent financial buyer, and a defined block of distressed supply between $4,470 and $5,000.

Scenarios and Targets: $4,120 Downside, $4,645 Base, $5,000 by Early 2027

The base case is continued range trade between $4,336 and $4,470 through the September FOMC, resolving higher into the fourth quarter. This requires the September meeting to deliver a hold, energy prices to stay soft, and ETF flows to turn marginally positive. Under those conditions gold clears $4,470, works toward $4,520, and tests the $4,645 August ceiling. Target: $4,645.

The bullish case requires the labor market to deteriorate further and force the Fed toward easing. July payrolls at minus 23,000 with continuing claims rising 24,000 is the seed of that scenario. A second negative payroll print, or a September hold accompanied by dovish guidance, would collapse real yields and bring the Western ETF buyer back. Holdings returning above the February record of 4,176 tonnes would confirm it. Sequential targets run $4,645, then $4,900, then $5,000 by early 2027, with longer-dated projections clustering near $5,033 by year-end 2026 and above $5,150 by December on the more aggressive models. Target: $5,000.

The bearish case begins with a September hike. Odds sit near 40% for that meeting, better than 53% for October and 73% for December — meaning a tightening move this year is closer to base case than tail risk. A hike would push real yields higher, restart ETF redemptions, and take gold through $4,336 toward $4,224 and then $4,120 to $4,084. Below that, $4,046 and $4,000 come into play, with the physical fair-value floor near $3,895. Target: $4,120, with $3,895 as the capitulation objective.

The wildcard is Hormuz, and it is genuinely two-sided. A negotiated reopening accelerates the energy disinflation already visible in July PPI, brings Fed easing forward, and supports the bullish case despite removing the safe-haven premium. An escalation spikes crude, revives the inflation constraint that drove the entire 2026 decline, and supports the bearish case despite the geopolitical bid.

Structural support does not depend on any of this. Central banks buying 850 tonnes annually, gold at 27% of official reserves against 22% for Treasuries, deficits at 6% to 7% of GDP, and 95% of surveyed reserve managers expecting official holdings to rise establish a demand base that operates independent of the rate cycle.

Verdict: The Floor Is Structural, the Ceiling Is the Fed, and Nothing Resolves Before September 16

Gold at $4,374 has recovered nearly 10% in a month and remains 22% below the January 29 high of $5,595.46. It is up 30.40% year over year and 1.92% year to date. Those three numbers describe a market that gave back a full year of gains in one quarter and has spent the following quarter rebuilding roughly a third of them.

The 2026 decline had one cause. Real yields rose because the Hormuz blockade drove oil above $100, which pushed CPI to 3.3% in March, which removed any path to Fed cuts, which triggered $12.7 billion of North American ETF redemptions in a single month and produced a record $611 monthly loss. Everything else — positioning, momentum, dollar strength at a 13-month high — amplified that mechanism rather than initiating it.

That mechanism is now partially reversing. July PPI printed 0.0% with energy down 3.1% and gasoline off 5.7%. CPI eased to 3.4% headline and 2.5% core, the slowest core reading since March 2021. September hike odds have fallen from 68% in June to roughly 40% now. Brent is at $87.04 and falling on OPEC and IEA demand downgrades. Each of those is a direct input into the real yield that governs the gold price.

What has not reversed is the level. The ten-year sits at 4.67%, the thirty-year above 5%, and October and December hike odds run 53% and 73%. Core PPI ex-trade accelerated to 0.4% from 0.1%, and services costs are rising at every production stage. Cool headline inflation with hot core services is the configuration that keeps a central bank restrictive without giving gold an inflation narrative to trade.

The structural side has never wavered. Central banks added 244 tonnes in the first quarter and are on pace for 850 this year. Gold passed Treasuries as the world's largest reserve asset at 27% of holdings against 22%. A record 45% of surveyed reserve managers plan to add to their own reserves. ETF holdings at 4,121 tonnes sit 1.3% below an all-time record despite a 28% price collapse — the financial buyer stepped aside, it did not liquidate.

Silver at $64.48 after a $67.06 print and a gold/silver ratio at 66 is the tell worth watching daily. Compression through 60 confirms the complex has genuine bid. Expansion above 70 means the industrial component is failing and gold retests the range low.

Base case $4,645. Bull case $5,000 into early 2027 on a labor-driven Fed pivot. Bear case $4,120 on a September hike, with $3,895 as the physical floor. The decision date is September 16, and until then this is a $4,336 to $4,470 range with a structural floor that has already been tested at $4,046 and held.

That's TradingNEWS