Gold, XAU/USD ($4,130) Clears $4,100 Following the Fed's 9-3 Hold at 3.50%-3.75%

Gold is trading around $4,130 an ounce by mid-morning in New York, having pushed through $4,100 for the second consecutive session. Spot closed Wednesday at $4,101.99, up 1.9% on the day after erasing earlier losses, with an intraday high of $4,116.26 that marked the strongest level since July 23. August COMEX futures settled about 0.1% lower at $4,036.30 before the Fed statement landed, then opened Thursday at $4,060.70 — up 0.6% — and have extended toward the $4,155 area as the session has developed.

The gap above the market is the number that defines this forecast. Spot gold set an intraday record of $5,595.47 on January 29, 2026, the culmination of a 2025 that delivered 53 separate all-time highs, a gain of more than 60%, and the best annual performance since 1979. From that peak, bullion has shed roughly 26%. It is still up approximately 24% year over year and up 1.23% over the past month, and it remains inside a 52-week range spanning roughly $3,250 to $5,595 — a band so wide it tells you more about the volatility of the last twelve months than about any current equilibrium.

The recent floor was set in mid-July at $3,975, a nine-month low and the first sustained break beneath $4,000 since the autumn of 2025. From there gold has built a jagged recovery: it opened near $4,130 on July 22, spiked to a cycle high near $4,160 the following day, surrendered nearly all of it on July 24 with a drop toward $4,000, recovered into the low $4,100s by July 26, then slid back toward $4,000 into the Fed meeting. That is four full round-trips inside a $160 band in ten sessions. The market is not trending. It is oscillating around a level it cannot decide is fair.

That level is roughly where the industry's own valuation work puts it. The Gold Valuation Framework maintained by the sector's global body places fair value at approximately $4,100 with a ±5% tolerance band, which brackets the entire July range. Gold is trading at fair value after a 26% drawdown from a peak that was, by any reasonable reading, a blow-off. The question for the second half is not whether the January high was an overshoot — it was. It is whether the structural bid that produced it survived the correction. Today's demand data gives a partial answer, and it is more constructive than the price action suggests.

The Nine-Three Hold Delivered a 1.9% Day and Not Much More

The Federal Open Market Committee left the target range at 3.50%–3.75% on Wednesday for a fifth consecutive meeting, on a 9–3 vote. The Cleveland, Minneapolis and Dallas presidents all dissented in favor of a quarter-point increase. Pre-meeting pricing had assigned roughly 40% odds to a hike, with one probability feed showing 66.3% for a hold — unusually wide uncertainty, and unusual in that the tail risk pointed toward tightening rather than easing. The removal of an immediate hike was the entire basis for gold's 1.9% move.

The chair's press conference capped it almost immediately. He declined to offer forward guidance, said the committee needs to observe market reaction to developments direct and unfiltered, and reiterated the commitment to returning inflation to the 2% target. Most consequentially for bullion, he stated that if inflation remains elevated throughout the forecast period, higher interest rates could become an appropriate policy response. That is not a hawkish surprise — it is a hawkish standing condition, and it means every subsequent inflation print carries a hike premium that gold has to absorb.

The market's own pricing reflects it. One feed put the probability of a September increase at roughly 80% before the meeting, and the consensus after the decision was that hike pricing had been pushed forward rather than removed. September is now the live meeting, and the incoming inflation data between now and then is the only input that matters. For an asset whose defining characteristic is that it pays nothing, a central bank that has explicitly reserved the right to tighten further is a persistent headwind regardless of what it did on any given Wednesday.

The honest read on the 1.9% move is that it was relief, not repricing. Gold went into the meeting near $4,020, hovering close to its lowest level since November 2025, having spent the previous week defending $4,000. A hold that removed a 40% tail risk produced a bounce back to fair value and stopped there. That is a market pricing the absence of bad news rather than the arrival of good news. The distinction matters for the forecast, because it means the level of support currently visible around $4,000 to $4,020 is being held by the absence of a catalyst rather than by active accumulation from price-insensitive buyers. Those buyers exist, but they operate on a quarterly clock rather than a daily one.

The Long End at 5.21% Is the Position Gold Has to Fight

The bond market's response to the hold was the genuinely important market event, and it cut directly against bullion. The 30-year Treasury yield surged twelve basis points to 5.21%, the highest level since 2007 and a nineteen-year peak. The 10-year rose more than seven basis points to 4.677% before easing to around 4.65% on Thursday. The two-year fell four basis points to roughly 4.24%. That combination is a textbook bear steepener: traders removed the near-term hike from the front end while demanding materially more compensation at the long end for inflation risk they believe the committee is under-addressing.

For a zero-yield asset, this is the binding constraint. Gold competes against the real return available on Treasuries, and a nominal 5.21% at the long end against core inflation running at 3.3% implies a real yield near 1.9% — a level at which the opportunity cost of holding bullion is meaningful rather than theoretical. Every basis point at the back end is a direct charge against the case for owning an ounce. The entire 2024–2025 advance ran on the expectation of falling real yields. That expectation has been fully priced out.

The chair's own framing amplifies the mechanism. By noting that tightening in financial markets is already performing part of the Fed's work, and by declining to guide, he has effectively deputized the bond market as the policy instrument. That means inflation surprises now transmit straight into the long end without a communication layer to absorb them, and from the long end straight into gold's opportunity cost. It also means the volatility of that transmission is higher than it would be under a guiding central bank, which is part of why bullion has been round-tripping $160 bands in a fortnight.

There is a second-order argument running the other way and it deserves weight. A curve steepening driven by term premium and long-run inflation compensation — rather than by growth optimism — is historically a constructive environment for gold over multi-year horizons, because it signals eroding confidence in the fiscal and monetary anchor. The reserve-composition data supports that reading: gold reached 27% of global official reserves at the end of 2025, surpassing US Treasuries at 22% for the first time. That is the structural story. It does not pay the carry between now and September.

Core PCE at 3.3%, GDP at 1.5%: The Data Cut Both Ways

Thursday's eight-thirty release was the second half of the week's verdict and it landed marginally in gold's favor. The June personal consumption expenditures price index fell 0.1% on the month, bringing the annual rate to 3.7% from 4.1% in May. Core PCE rose 0.1% against a 0.2% forecast, with the year-over-year figure easing to 3.3% from 3.4% — a reading that had been a three-year high. Energy goods and services prices tumbled 5.9%, with gasoline down 9.2% during the brief Middle East lull. Housing inflation moderated to 0.2%.

Second-quarter GDP grew at a 1.5% annual rate, below the 1.8% consensus and down from 2.1%. The composition was better than the headline suggests: consumer spending accelerated to a 3.2% annual clip from 0.5%, the underlying-strength measure that strips out government and trade expanded at 3.9%, and business investment excluding housing rose 8.4%. Imports and a 0.7% inventory drawdown produced the miss. Jobless claims came in at 197,000 for the week ended July 25, up 9,000 and still historically low.

This is, on paper, the configuration gold is supposed to own: growth decelerating toward 1.5%, inflation stuck 130 basis points above target, and a central bank that cannot ease without abandoning its mandate. Slowing growth with sticky prices is the textbook environment for a non-yielding real asset. The reason it is not producing a larger move is that the market is reading the same data through the rate channel rather than the stagflation channel, and through that lens a 3.3% core print keeps September live.

The deeper problem is that June's improvement is already unwinding. The gasoline decline that flattered the print came during a ceasefire window that has since closed — crude jumped 6.6% on Wednesday, with Brent settling at $90.74 before easing to $88.93. On a quarterly basis the PCE index surged 5.1% headline and 3.4% core, meaning the monthly relief disappears entirely when the frequency is widened. If energy pushes the August and September readings back up, the hike returns to the table, real yields rise further, and gold retests $4,000. That risk is the single largest downside variable in this forecast, and it is not a tail — it is the base-case path if crude holds above $90.

What Actually Broke the Rally: The War That Turned Bullish Into Bearish

The mechanism behind this year's 26% correction deserves precise statement, because most narratives get it backwards. The US–Iran conflict that escalated in late February 2026 was, counterintuitively, bearish for gold. It should have been the perfect safe-haven catalyst. Instead, rising oil prices supercharged inflation expectations, which prompted markets to price out Federal Reserve cuts and eventually assign meaningful probability to hikes. Higher-for-longer real yields strengthened the dollar. Those are gold's twin headwinds, and the war delivered both simultaneously.

The damage was concentrated and fast. Gold shed more than 10% in March alone, its worst monthly decline since June 2013. The full second quarter produced the steepest quarterly decline in thirteen years. Inflation ran to 4.2% year over year by May, which is precisely the reading that should support bullion — and instead it destroyed it, because the transmission ran through policy expectations rather than through purchasing-power fear. Gold broke decisively below the $4,300 support level following renewed US strikes on Iran and the rising probability of a September hike.

This is the analytical point that governs the rest of the year: gold's relationship to geopolitical risk is conditional on how that risk transmits to real rates. A supply shock that raises energy prices in an environment where the central bank is already above target and openly considering tightening is a net negative for bullion, because the rate response dominates the haven bid. The same shock in an easing cycle would have added several hundred dollars. Positioning has not fully absorbed this distinction, which is why every escalation headline still produces an initial spike that fades within a session.

The corollary is the constructive scenario. If oil stabilizes or retreats — Brent has already slipped below $90 from Wednesday's $90.74 settle, with WTI at $83.09 — the inflation impulse fades, September hike pricing compresses, real yields fall, and the same geopolitical backdrop flips from headwind to tailwind without any change in the underlying conflict. Watch the crude tape more closely than the Middle East headlines. Bullion is not currently trading the war; it is trading the war's second derivative through the oil complex into the front end of the rate curve.

The Chart: $4,021 Floor, $4,132 Ceiling, $3,975 the Line That Matters

The technical structure is a symmetrical triangle that is running out of room. Gold has recovered off support around $4,021, with recent bullish candlesticks indicating buyers re-entering after several sessions of consolidation. The pair remains above that floor but requires a decisive break above $4,132 for anything more significant. That level is doing double duty: it marks the convergence of the upper triangle boundary and the 200-period exponential moving average, which makes it the single most important price on the chart today.

Spot is sitting essentially on it. A close above $4,132 that holds through the New York session would be the first genuine technical improvement since the January breakdown, and it would open the path toward the $4,157 to $4,160 zone that capped the July 23 spike. Above that, $4,236 marks the upper boundary of the month's projected range, and $4,300 is the level lost after the renewed strikes on Iran — the first structurally meaningful resistance rather than merely technical.

Below the market, the ladder is well defined. Immediate support at $4,021 is the triangle floor, followed by the $4,000 psychological level that has now been defended on multiple attempts including at least one breach. Stronger downside confirmation comes on a sustained move under $4,020. Beneath that sits the mid-July nine-month low at $3,975, and losing that opens genuine air — the next reference is the November 2025 zone, and several analysts have flagged scope for a further 16% decline toward $3,400 to $3,500 if the September hike materializes.

The broader moving average structure remains bearish and has not repaired. Multiple months of declines have left the medium-term averages sloping down and above price on most timeframes, which is why the recent rebound has been characterized as fading downside momentum rather than a trend change. The day's projected range of $4,007.83 to $4,157.41 is wide enough that standard intraday assumptions should be treated with caution, particularly given that a $4,000 to $4,160 swing has already occurred inside a single week this month. The disciplined framing: neutral to mildly bearish below $4,080, constructive above $4,132, and structurally broken below $3,975.

Q2 Demand: Central Banks Bought 289 Tonnes While ETFs Sold Forty-Five

The quarterly demand data published today is the most substantive input available for the second-half forecast, and the headline divergence is stark. Gold-backed exchange-traded funds came under selling pressure with net outflows of 45 tonnes across the second quarter. Central banks, over the same three months, bought 289 tonnes. One cohort was liquidating into the steepest quarterly price decline in thirteen years. The other was accumulating into it.

The ETF outflows were described as moderate and were driven by exactly the mechanism outlined above: weaker gold prices combined with upward revisions to both inflation and interest rate expectations, particularly in North America, alongside a strengthening dollar. That is a coherent, rational, rate-driven exit. It is also, critically, a Western phenomenon rather than a global one. The composition of that selling matters enormously for what happens next.

Central bank buying is the other side and it recovered sharply. After a notable first-quarter slowdown that followed a downward data revision, official-sector purchases returned to the elevated levels typical of the last four years. The People's Bank of China added 14.93 tonnes in June alone — its twentieth consecutive month of purchases and its largest single-month addition since 2023 — buying directly into a historic quarterly decline. Poland added 14 tonnes in April, bringing its year-to-date total to 45 tonnes. These are price-insensitive reserve managers executing multi-year mandates, not tactical allocators.

Bar and coin investment held steady year over year at 307 tonnes, signalling a return to more typical buying after two extraordinarily strong quarters. Jewellery demand fell to 278 tonnes, the lowest quarterly volume since the pandemic, as high prices and inflationary pressure constrained affordability — though spending on gold jewellery rose 14% year over year to $40 billion, which confirms the metal's share of wallet even as tonnage falls. Total supply held steady at 1,269 tonnes, with a 2% increase in mine production offsetting a 6% decline in recycling as lower prices discouraged the sale of old jewellery. The forward guidance from the same work: investment should remain the primary driver through the second half, supported increasingly by over-the-counter activity and Asian buying, with central banks on course for another strong year though likely below 2025.

The 298-Tonne Overhang: Why ETF Holders Are the Ceiling, Not the Floor

The single most underappreciated constraint on a gold recovery is the volume of exchange-traded fund holdings sitting underwater. At prices around $4,000, approximately 298 tonnes of gold held inside ETFs is at a loss — up from 270 tonnes when the metal was still above $4,250. That is not a static figure. It expands as price falls and contracts as price recovers, which creates a specific and predictable dynamic on the way back up.

Those holders are not long-term allocators. They are traders waiting for an exit, and their entry points cluster in the $4,250 to $5,000 band established during the late-2025 and January melt-up. Every dollar of recovery toward those levels converts a paper loss into a break-even and generates a redemption decision. The mechanical result is that a gold rally toward $4,500 will meet supply that a rally toward $4,150 will not, because the pool of break-even sellers grows as price climbs. That is a structural ceiling on the near-term advance rather than a floor beneath it.

The flow data through the second quarter bears this out. Gold-backed funds recorded net outflows of 16 tonnes in May and continued bleeding into the first half of June before a $1.1 billion weekly inflow snapped four straight weeks of redemptions. That snapback was encouraging and it did not change the aggregate: the quarter still finished at negative 45 tonnes. Selling pressure easing is not the same as buying pressure arriving, and the distinction has been repeatedly blurred in commentary this month.

The forward guidance is explicit on this point. For North American funds, the principal near-term headwind is the opportunity cost of holding gold, and higher real yields combined with shifting policy expectations could restrain third-quarter flows. Flows in North America and Europe are expected to be episodic rather than trending. One major bank has already cut both its December price forecast and its ETF demand projections, while another has said its $5,200 second-half target is now increasingly dependent on a revival in ETF demand and evidence that lower oil prices are genuinely improving the case for dovish policy. Read that as the sell side conceding that the Western investment bid is the binding variable, and it is currently absent.

The Sovereign Bid: China's Twentieth Straight Month and 27% of Reserves

The counterweight to Western fund liquidation is official-sector accumulation, and it is operating on a completely different logic. The People's Bank of China has now bought for twenty consecutive months, with June's 14.93 tonnes its largest single-month addition since 2023 — executed while gold sat roughly 28% below its January peak. Poland has added 45 tonnes year to date. The Czech central bank has been a consistent if smaller buyer. Second-quarter official-sector purchases of 289 tonnes followed 244 tonnes in the first quarter, both above the five-year average.

The driver is reserve composition, not price. The shift began in earnest in 2022 when roughly $300 billion of Russian foreign exchange reserves were frozen, demonstrating that paper claims held abroad can be immobilized overnight while physical metal cannot. The resulting reallocation has been steady and directional. BRICS-plus nations now hold 17.4% of global gold reserves, up from 11.2% in 2019. Most strikingly, gold reached 27% of global official reserves at the end of 2025, surpassing US Treasuries at 22% for the first time in the modern era.

That crossover is the structural story and it is not a price forecast. Central banks buying roughly 1,000 tonnes annually against total supply of approximately 1,269 tonnes per quarter means the official sector is absorbing something in the range of 20% of all gold reaching the market. That is a permanent removal of float — coins that do not return to the market on a rally and do not get liquidated on a drawdown. It raises the long-run floor and dampens the depth of corrections. It does not, on its own, produce rallies, because the buying is calibrated to reserve targets rather than to momentum.

The forward view holds that central banks remain on course for another strong year, though likely below 2025's exceptional pace. A survey of the official sector found 95% of central banks expected global gold holdings to rise over the following twelve months, with not a single respondent expecting a decline. The practical implication for the forecast: sovereign demand justifies a structurally higher price floor than the pre-2022 regime would imply, which is why the $3,975 low held and why calls for $3,000 look unserious. It does not justify a return to $5,595 without the Western investment bid returning alongside it.

Jewellery at a Pandemic Low and What Asia Is Actually Doing

Jewellery demand fell to 278 tonnes in the second quarter, the weakest quarterly volume since the pandemic, as elevated prices and broader inflationary pressure continued to constrain affordability. That is the price-elastic portion of the market functioning exactly as expected — at $4,100 an ounce, physical consumption contracts. High prices continue to reduce jewellery demand in the two markets that matter most, India and China, and volumes are expected to remain under pressure through the second half.

The value figure tells a different and more useful story. Spending on gold jewellery rose 14% year over year to $40 billion, which confirms that gold's share of consumer wallet is expanding even as the tonnage it buys shrinks. Consumers are not abandoning the metal; they are getting less of it for more money. That is a demand curve with meaningful price elasticity in volume and considerable inelasticity in spend, and it means jewellery acts as a shock absorber rather than a swing factor — it cushions drawdowns by absorbing supply at lower prices and steps aside during rallies.

The Chinese import data is the more actionable input. Imports jumped to approximately 173 tonnes in June, the strongest monthly inflow since March 2024, bringing first-half totals to roughly 820 tonnes. That surge coincided with softer international prices, a stronger yuan, and the deployment of bank import quotas. It is a textbook demonstration of Asian buyers behaving as price-sensitive accumulators: Western funds sold the correction and Chinese importers bought it. The forward guidance explicitly names Asian investment and over-the-counter activity as increasingly important drivers of demand growth through the second half.

Bar and coin demand at 307 tonnes rounds out the physical picture, holding steady year over year after two exceptionally strong quarters. The characterization was a return to typical levels rather than a collapse, and it is expected to moderate further after a strong start to the year. Taken together, the physical market — jewellery, bars, coins, and Asian imports — is doing what it always does in a correction: absorbing metal quietly at lower prices while the headlines focus on fund redemptions. That absorption is why $3,975 held. It is also why the recovery from that low has been slow and grinding rather than sharp.

Supply Is Not the Constraint: 1,269 Tonnes and a Recycling Squeeze

Total gold supply held steady at 1,269 tonnes in the second quarter, which removes the supply side from the forecast almost entirely. A 2% year-over-year increase in mine production was offset by a 6% decline in recycling, as lower quarter-over-quarter prices discouraged the sale of old jewellery back into the market. That is the recycling channel functioning as a natural stabilizer — it expands into rallies and contracts into drawdowns, dampening both.

The recycling contraction is worth dwelling on because it is a genuine second-half positive that gets almost no attention. Scrap supply is the most price-elastic source of metal in the market, and its 6% decline at $4,100 tells you the marginal seller of existing above-ground stock has stepped back. Combined with the 298 tonnes of ETF gold sitting underwater — holders who will sell into a rally rather than into a decline — the available float is currently concentrated in weak hands positioned above the market rather than below it.

Mine production offers only modest growth potential from here, and the forward guidance says as much for both mine output and recycling. This is a structural feature of the industry rather than a cyclical one: new supply requires eight to twelve years from discovery to production, permitting timelines have lengthened, and grade profiles at existing operations continue to decline. Gold's supply curve is close to vertical over any horizon shorter than a decade, which means the price is set entirely by the demand side of the ledger.

The cost side is where the second half gets interesting for the producers rather than the metal. All-in sustaining costs across the sector have been running well below spot, and the sharp move lower in energy prices during June's lull works directly through diesel, power and reagent costs into mining margins. Management commentary on sustaining costs during the current earnings season will be the key input for third-quarter margin forecasts. For the metal itself, none of this matters much — supply that cannot respond to price within a decade does not cap a rally. For the equities, it is the entire story, and it is why the miners currently offer a materially different risk profile than the bullion does.

Miners Took a Twenty-One Percent Beating and Now Carry the Leverage

The gold equity complex absorbed considerably more damage than the metal did. The senior miners ETF shed 21% in the second quarter, sliding from $96 in early April to roughly $75 by June 30 — one of the ugliest three-month stretches for the sector in more than a decade, against a metal decline of roughly a quarter from the January peak. The fund remains up close to 50% over the trailing year, which is the gap contrarians hunt: acute recent pain sitting on top of an intact longer trend.

July compounded it. The sector bellwether shed 14.9% over the month, contributing a 1.58 percentage point drag to the index. Barrick fell 13.7% for a 1.09 point drag on an 8.0% weight. Kinross dropped 21.7%. The largest streaming company lost 15.3%, and AngloGold Ashanti fell 16.5%. Those are declines two to three times the magnitude of the metal's move over the same window, which is operational and financial leverage working in the direction nobody wants it to.

The technical picture confirms the damage. The 50-day moving average crossed below the 200-day on June 26, a long-term bearish signal, and momentum turned negative in the second week of July. The sector is oversold on most measures and has not yet produced the reversal signal that would confirm a bottom. Anyone treating the miners as a leveraged entry into a gold recovery needs to accept that the leverage has been running against them for four months and that a metal move back to $4,300 would need to be sustained rather than spiked to repair the equity charts.

The offsetting case is margin arithmetic. Sustaining costs across the senior producers have historically run well below $1,500 an ounce, which at $4,130 spot implies gross margins that remain historically exceptional even after a 26% price decline. The energy price collapse during the June lull is a direct tailwind to those costs, and it arrives just as the earnings season lands. If operational execution holds and cost guidance improves, the equities are priced for a metal considerably below $4,000. That is the asymmetry: the miners have already discounted a bear case for gold that the physical demand data does not currently support.

The Forecast: $4,500 Base, $4,900 Bull, $3,750 Bear Into Year-End

The base case, at roughly even odds, is consolidation between $4,000 and $4,500 through the third quarter with a grind higher into the fourth. This requires the Fed to hold in September, oil to stabilize rather than accelerate, and ETF flows to stop bleeding without necessarily turning positive. Under this path, gold works through the $4,132 triangle-and-200-EMA convergence in coming sessions, tests $4,236 and then the $4,300 level lost after the renewed Iran strikes, and finishes the year in the $4,400 to $4,500 range. That is aligned with the more conservative sell-side target of $4,500 for the fourth quarter and sits inside the ±5% band around $4,100 fair value for most of the interim.

The bull case requires one specific thing: a revival in Western ETF demand. That is the explicit condition attached to the most aggressive published target of $5,200 for the second half, and it is the same variable behind the $4,900 year-end call that was cut in June. The trigger sequence is identifiable — Brent sustained below $85, core PCE printing 3.1% or lower in August, September hike pricing collapsing, and the 30-year retreating from 5.21% toward 4.90%. Get all four and the 298-tonne underwater overhang converts from resistance into a squeeze, because those holders stop being sellers and momentum allocators return. Assign this perhaps 25%.

The bear case is a September hike, and it has a clean path. Crude holding above $90 rebuilds the energy contribution to August and September inflation, hike pricing firms toward the 80% level already seen once this month, the 30-year pushes through 5.25%, and real yields move decisively above 2%. Gold loses $4,021, then $4,000, then the July low at $3,975. Below that the market has flagged scope for a further 16% decline toward $3,400 to $3,500. Weight this around 25% and note that it is driven entirely by the oil complex, not by anything in the demand data.

The disciplined posture at $4,130 is constructive accumulation on weakness toward $4,000 rather than chasing strength into $4,150. Central banks bought 289 tonnes into the worst quarter in thirteen years, recycling supply is contracting, Asian imports are running at two-year highs, and gold now exceeds Treasuries as a share of global official reserves. The structural floor is higher than the price action suggests. The ceiling is the ETF overhang and the September meeting, in that order.

That's TradingNEWS