Gold (XAU/USD) at $4,389 Faces $4,500 Test as Central Banks Absorb 289 Tonnes and the 30-Year Yield Hits 5.3%

Gold spot traded $4,389.83 Tuesday, down $26.86 or 0.61% against Monday's $4,416.75 close. The session opened at $4,416.75, ran to $4,436.15, and got sold back to a low of $4,386.10 — a $50.05 range and the second consecutive rejection at the same overhead level. December COMEX futures printed $4,446.60, down $27.10 or 0.61%.

The rejection point is the story. Gold tagged $4,435 last week and failed. It tagged $4,436.15 Tuesday and failed again. Two identical rejections inside four sessions at a level that also marks the metal's two-month high establishes $4,435 to $4,450 as the operative ceiling until it breaks on a closing basis.

The move underneath it has been real. Gold rose 10.52% over the past month and is up 33.53% year-over-year. It posted its largest one-week gain since January during the first week of August, then extended to a fresh two-month peak near $4,450 before correcting to end that week virtually unchanged. Monday delivered a 0.50% advance, the second straight positive session, carrying the metal above $4,400 for the first time in eight weeks.

Silver moved with it, up 0.47% and trading near $65.19. Platinum added 0.13%. The precious complex is running as a single trade, and that trade is a Fed repricing rather than a safe-haven bid.

Year-to-date performance tells the harder truth. Gold is up 1.24% in 2026. The metal opened the year near $4,331, ripped to an all-time high of $5,602.23 on January 29, then bled for six months to a July trough at $4,053.11. Everything since has been recovery off a base, not continuation of a trend. At $4,389.83, gold sits 21.6% below its record and 5.4% above the yearly open.

July closed with a 0.5% gain, the first monthly increase since February. Five consecutive losing months preceded it. The metal is one good week into repairing a structural breakdown that took half a year to build, and it is being asked to do that against a 30-year Treasury yield at a nineteen-year high.

The 52-week range runs $3,351.33 to $5,602.23 — a $2,250.90 band, or 51.3% of the current price.

The Rate Repricing Is Doing All the Work: September Hold Odds at 69.9%

Gold's entire August recovery traces to one shift: the market has stopped pricing a Federal Reserve rate hike.

CME Group data puts the probability of the Fed holding at 3.50%–3.75% in September at 69.9%. Hike odds sit near 35%, down from close to 50% before last week's data and down from a majority reading earlier in August. Markets are no longer fully pricing a rate increase by year-end, a complete reversal from where positioning sat seven days ago.

The data drove it. July retail sales fell 0.6% against expectations for a 0.1% gain — a 70 basis point downside surprise. Excluding autos, sales fell 0.3% versus a forecast 0.2% gain. The University of Michigan Index of Consumer Sentiment dropped to 51.0 in August from 55.2 in July, sitting roughly six points above the historic low of 44.8 set in May. July inflation printed tame. July housing starts collapsed 12.4% to 1.239 million against a 1.35 million forecast, with single-family starts down 9.9% to 808,000.

That sequence — soft inflation, weak consumption, contracting construction — removes the case for tightening into September. Every basis point of hike probability that comes out of the curve lowers the opportunity cost of holding a non-yielding asset, and gold captures it directly.

The problem is that this trade has now largely happened. From roughly 50% hike odds to roughly 35% produced a move from $4,053.11 in late July to $4,436.15 Tuesday — a $383 advance, or 9.4%. Extracting further upside from the same mechanism requires hike odds to keep falling toward zero, and 69.9% hold probability leaves limited room.

Fed Chair Kevin Warsh has reaffirmed commitment to reducing inflation without signaling near-term increases in borrowing costs. The July FOMC minutes land Wednesday, August 19, and his Jackson Hole address follows. Those two events carry the balance of the rate story for August, and the risk is asymmetric: dovish confirmation is already in the price, hawkish surprise is not.

The next hard data point is the August CPI release on September 10, followed by the FOMC meeting.

The Contradiction Gold Cannot Resolve: 30-Year Treasuries at 5.323%

Gold rallying while the long end of the Treasury curve breaks to nineteen-year highs is the single most difficult fact on this chart, and it should not be dismissed as noise.

The 30-year Treasury yield printed 5.323% Tuesday, up two basis points and the highest level since 2007. The 10-year hit 4.72% to 4.73%. Long-dated sovereign yields rose across every major market simultaneously — this is a global term-premium event, not a US-specific repricing.

Rising nominal and real yields are the textbook headwind for bullion. A non-interest-bearing asset competes directly against risk-free duration, and every basis point at the long end raises the carrying cost of holding metal. When the 30-year crossed 5.2% for the first time since June 2007 earlier this month, that dynamic was correctly identified as the primary constraint on any recovery.

Gold is ignoring it, and the reason is that the front end and the long end are telling different stories. The market has repriced the policy rate lower for September while the term premium at the back of the curve expands on deficit financing, AI-related corporate issuance and an oil-driven inflation premium. Gold trades the policy rate. Duration trades the term premium. Both can move against each other when the curve steepens for fiscal rather than monetary reasons.

That distinction is also gold's structural bull case. A 30-year at 5.323% driven by sovereign supply rather than growth is a solvency signal, and solvency signals are what central banks buy metal against.

The vulnerability sits in the sequencing. If Wednesday's minutes show a broad hawkish bloc — and Cleveland Fed President Beth Hammack has publicly pushed for immediate increases, having dissented in June — the front end reprices back toward tightening while the long end stays elevated. Gold then loses its only supportive input and faces the full weight of a 5.32% discount rate.

Real yields are the transmission mechanism, and they have moved decisively against the metal all year. Gold's 21.6% drawdown from the January record is the direct output of positive real rates, and nothing in the current configuration has reversed that.

Brent at $91.76 Cuts Both Ways for Bullion

Brent crude reached approximately $91.76 per barrel Tuesday as hopes for an extended US-Iran ceasefire collapsed. West Texas Intermediate traded $84.39, up 0.78%, touching $85 intraday — a third consecutive session of gains. The US Strategic Petroleum Reserve sits at its lowest level since 1982, removing the release valve that capped prior spikes.

The geopolitical stack is dense. President Trump rejected extending the 60-day interim agreement with Iran that expired Monday, stated an intent to inflict further economic pain on Tehran, and threatened military action against Oman over Strait of Hormuz access. A senior Iranian official responded that the country would shift to a fully offensive posture if diplomacy fails. Israel launched fresh strikes on Lebanon over the weekend. New US sanctions aimed at forcing Iranian capitulation are in preparation.

That should be an unambiguous gold bid. It is not, and the reason is the second-order effect. Crude at $91.76 lifts inflation expectations, inflation expectations lift the long end, and the long end raises the opportunity cost of holding metal. Gold's safe-haven demand from Middle East escalation is being partially neutralized by the rate consequence of the same escalation.

The evidence sits in the price action. Gold traded $4,446.60 in futures Monday while equities fell 0.52% and the VIX rose. A genuine flight to safety produces bullion outperformance against a falling S&P 500, and Tuesday delivered the opposite: gold down 0.61% with S&P 500 futures down 0.51% and Nasdaq-100 futures down 1.31%.

That correlation says the marginal buyer is trading the Fed, not the war. Gold has now spent the better part of 2026 failing to convert an active US-Iran conflict into sustained upside — the metal fell from $5,602.23 to $4,053.11 with that conflict running the entire time.

Middle East escalation has become the background condition rather than the catalyst. Traders have marked it to the curve and moved on. It takes an actual supply disruption through Hormuz, not another threat, to reprice bullion on geopolitics alone.

The Dollar Broke Down to 99.29 and Gold Took the Handoff

The Dollar Index fell to 99.29 early Monday, its lowest level since June 5, breaking the bullish trendline that defined the ascent from the January low at 95.55. That is a technical breakdown in the denominator of every gold quote, and it is the second pillar of the August recovery alongside the rate repricing.

The mechanism is direct. Gold is priced in dollars and held globally. A weaker dollar makes bullion cheaper for every non-dollar buyer simultaneously, and it removes the competing store-of-value bid that pulls reserve capital into Treasuries. The move from $4,053.11 in late July to $4,436.15 Tuesday coincided almost exactly with DXY sliding from its summer highs to a two-and-a-half-month low.

The dollar fell for the same reason gold rose: soft US data collapsed the tightening case. Weak retail sales, deteriorating sentiment and tame inflation removed the rate differential that had supported the currency, and the trendline broke.

The complication is what a falling dollar alongside a 5.323% 30-year actually represents. A currency weakening while long-dated yields hit nineteen-year highs is not a liquidity expansion. It is foreign capital demanding higher compensation to hold US duration while simultaneously reducing exposure to the currency — a term-premium and credibility event rather than a monetary easing.

That configuration is historically the most bullish backdrop gold gets. Reserve managers reallocating away from dollar assets have exactly one liquid, credit-risk-free alternative, and it is metal. The World Gold Council's 2026 survey found 89% of global reserve managers expect central bank gold holdings to increase over the next twelve months.

Gold's failure to convert a clean dollar breakdown into more than a two-session, $383 advance is therefore a demand statement about the Western investor cohort, not about sovereign buyers. Those two groups are moving in opposite directions and have been all year.

Currency dynamics also matter to the physical market. A positive local price spread in Shanghai has driven Chinese import volumes throughout 2026, and dollar weakness widens that arbitrage.

Twenty-One Straight Months: PBoC Adds 20 Tonnes, Holdings Reach 2,366

The People's Bank of China added 20 tonnes of gold in July, its largest monthly purchase since October 2023, lifting official holdings to 2,366 tonnes. That extends the buying streak to 21 consecutive months and brings Chinese reserves to 76.08 million ounces.

The acceleration is the point. The PBoC added 8 tonnes in April, 10 tonnes in May, 15 tonnes in June and 20 tonnes in July. Each monthly print exceeded the prior one, and each was described at the time as the largest since December 2024 until the next one topped it. First-half 2026 purchases totaled 40 tonnes. Total accumulation over the past 20 months reached 82 tonnes.

Compare that to the full-year 2025 figure of 27 tonnes. China bought more gold in the first six months of 2026 than in all of the preceding year, and did so while the price fell from $5,602.23 to $4,053.11. That is dip-buying by a sovereign balance sheet with a thirty-year mandate, executed against a market where futures traders were liquidating positions into positive real rates.

Gold now represents roughly 9% of China's $3.8 trillion in foreign exchange reserves. That share has climbed steadily but remains far below Western peer allocations, which is the structural argument for continuation: the gap itself is the demand forecast.

The official number understates the reality. Chinese gold imports reached approximately 163 tonnes in May while reported PBoC purchases for the same month were 9.95 tonnes. The difference flows into commercial bank vaults, Shanghai Gold Exchange settlement, jewellery and industrial demand, and potentially into accumulation programs outside official reserve reporting. Official PBoC data functions as a floor on Chinese sovereign accumulation, not a ceiling.

Physical demand inside China stayed weak in July. Gold withdrawals from the Shanghai Gold Exchange fell 8% month-on-month to 80 tonnes. Average daily volumes in Shanghai gold futures declined 4% to roughly 292 tonnes, though net long positioning rose as price volatility compressed.

The jewellery sector remains the persistent drag, with record prices continuing to suppress consumption volumes even as value-terms spending holds.

Central Banks Bought 289 Tonnes in Q2 — A Record for Any Second Quarter

Global central banks purchased a net 289 tonnes of gold in the second quarter of 2026, a 62% increase over the second quarter of 2025 and the highest volume ever recorded for any second quarter in the World Gold Council's historical data series. That followed 244 tonnes in the first quarter, up 17% quarter-over-quarter and above the five-year average.

Combined first-half sovereign accumulation reached 533 tonnes. Central banks have averaged approximately 1,000 tonnes per year since 2022, double the pace of the preceding decade.

The composition matters as much as the total. Poland led alongside China, adding 31 tonnes in the first quarter to push reserves to 582 tonnes against a stated 700-tonne target. Uzbekistan added 25 tonnes. Kazakhstan added 12. The Czech National Bank added 5. South Korea's central bank announced its first gold purchases in thirteen years and stated it would raise gold's share of foreign exchange reserves over the medium to long term, buying through US-listed gold ETFs.

That last detail is structurally significant. A sovereign reserve manager routing accumulation through listed ETF vehicles rather than physical allocation creates a demand channel that shows up directly in Western fund flows — the same flows that have been in redemption all year.

The buying happened into weakness, which is the entire argument. The average LBMA gold price for the second quarter settled at $4,506 per ounce, with prices declining approximately 14% over the period. May alone saw 16 tonnes of net ETF redemptions as Western investors managing against quarterly benchmarks exited on rising real yields.

Two cohorts, opposite directions, same asset. Western ETF demand is a real-yield function with a quarterly horizon. Sovereign demand is a reserve-composition function with a thirty-year horizon. A futures trader faces career consequences for holding through a 16% drawdown; a reserve manager faces none.

Turkey's liquidation of as much as 80 tonnes in March to defend the lira demonstrated that sovereign selling exists. Net official demand still turned decisively positive.

Chinese ETFs Flipped From a Record $2.2 Billion Outflow to Daily Inflows

Chinese gold-backed ETFs attracted RMB5 billion, or $744 million, in July, lifting total holdings 5 tonnes to 282 tonnes and raising assets under management 3% to RMB250 billion, roughly $37 billion. The inflows continued into August, with the funds recording positive daily flows on nearly every trading day through August 12.

That reverses one of the sharpest reversals in the product's history. Chinese gold ETFs lost RMB15 billion, or $2.2 billion, in June — the worst month on record. The outflow, combined with the falling price, cut total AUM 16% to RMB243 billion, the lowest level since December 2025, and reduced holdings by 17 tonnes to 277 tonnes.

Even with June's damage, first-half net inflows totaled RMB40 billion, or $5.6 billion, the second-strongest first half on record. Holdings rose 29 tonnes over the six months.

The behavior is regime-dependent rather than structural, and the switch is domestic equities. June's record outflow coincided with a Chinese equity rally that pulled retail capital out of metal and into stocks. July's reversal coincided with weaker equities, recurring geopolitical uncertainty and visible PBoC accumulation. Chinese gold ETF demand functions as a relative-attractiveness trade between two domestic asset classes, and it can flip inside a single month.

Institutional participation is rising as the market deepens. The Shanghai Gold Exchange has pledged closer cooperation between Hong Kong and Shanghai markets and expansion of its international board, which permits offshore investors to trade yuan-denominated gold. The clearing system's first cohort includes eleven major banks — five mainland Chinese lenders and six international institutions.

Full-year context tempers the enthusiasm. Domestic Chinese gold ETF holdings grew 28.677 tonnes in the first half of 2026, down 66.17% from the same period a year earlier, according to the China Gold Association. The absolute flow is positive. The growth rate collapsed.

SPDR Gold Shares alone holds in excess of $70 billion, dwarfing the Chinese complex in size. The Chinese funds matter for velocity and directional consistency, not scale.

China Imported 764 Tonnes in the First Half, Up 138%

China's net gold imports totaled 152 tonnes in June, a 2-tonne increase month-on-month and the highest monthly level since March 2024. First-half imports climbed 138% year-over-year to 764 tonnes.

That figure is the most underappreciated number in the gold market. Global annual mine production runs approximately 3,000 to 3,500 tonnes, with recycling adding another 1,000 to 1,500 tonnes. China absorbed 764 tonnes of physical metal in six months — roughly a fifth of global mine supply for the period — while the price fell 28% from its January peak.

The monthly progression through 2026 shows sustained intensity. March imports reached 143 tonnes, a 49% month-on-month rise that brought first-quarter net imports to 316 tonnes, up 182% quarter-over-quarter and 333% year-over-year. April delivered 157 tonnes, up 10% month-on-month and 40% year-over-year. May came in at 151 tonnes. June hit 152 tonnes.

Five consecutive months above 140 tonnes establishes this as a standing bid rather than opportunistic accumulation. The positive local price spread in Shanghai has encouraged importers throughout, and dollar weakness widens that arbitrage further.

The physical absorption is not going into jewellery. Chinese jewellery demand has declined for six consecutive years on a year-over-year basis as record prices curbed consumption, and wholesale demand stayed tepid through July with SGE withdrawals down 8% month-on-month to 80 tonnes. Bullion investment buying and official-sector accumulation are absorbing the volume.

That composition matters for price sensitivity. Jewellery demand is price-elastic and falls when gold rallies. Investment and reserve demand is price-inelastic or actively counter-cyclical — it accelerated as gold dropped from $5,602.23 to $4,053.11. The Chinese bid strengthens into weakness, which converts it into a floor mechanism rather than a momentum amplifier.

Chinese demand is why the $4,000 to $4,050 area held in July despite a technical breakdown that carried price below the 200-day moving average near $4,500 and pushed the metal under $4,000 intraday. That floor is the base the current recovery is built on.

The Drawdown Math: 21.6% Below the $5,602.23 January Record

Gold at $4,389.83 sits 21.6% below the all-time high of $5,602.23 set January 29, 2026, and 31.0% above the 52-week low at $3,351.33. Those two numbers frame everything.

The path was violent. Gold traded near $4,550 at the end of December 2025, opened January around $4,331, then ripped to $5,602.23 inside four weeks — a 29.3% advance in under a month. It then bled continuously through the first half, reaching $4,053.11 by the end of July. That is a $1,549.12 decline, or 27.7%, spread across six months against five consecutive losing months.

The catalyst for the reversal was the US-Iran conflict, which forced the Federal Reserve into a more hawkish posture and kept the policy rate at 3.75% through the period. Positive real interest rates did the rest. Gold futures traders liquidated into it, and the metal fell roughly 7% year-to-date by late July before the August recovery pulled it back to plus 1.24%.

The technical damage from that decline has not been repaired. Gold broke below its 200-day moving average around $4,500 in June, and that level now sits overhead as resistance. Price spent six weeks consolidating in a range straddling the $4,074 to $4,112 pivot zone, defined by the 61.8% retracement of the March decline, the March low, and the October high-week reversal close. The 2026 yearly open at $4,312 to $4,319 was reclaimed during the August advance and now functions as the first meaningful support shelf beneath current levels.

The one-year comparison flatters the metal. Gold is up 31.72% to 33.53% over twelve months because the starting point was $3,351.33 in August 2025, before the run that carried it above $5,600. Measured from the peak, 2026 has been a bear market with a recovery attached to the end of it.

The recovery has covered $336.72 from the July low to Tuesday's price, or 8.3%. It needs another $1,212.40 to make new highs.

Technical Structure: RSI at 65.37, Upper Bollinger at $4,485, Support at $4,360

The daily chart holds a bullish near-term bias with spot above both the 100-day simple moving average and the Bollinger middle band, keeping price supported inside the recovery structure. The Relative Strength Index at 65.37 leans toward overbought without reaching it, which reads as firm momentum that is vulnerable to consolidation after a 10.52% monthly advance.

Immediate resistance aligns with the upper Bollinger band near $4,485. Between spot and that level sit the twin rejections at $4,435 and $4,436.15 and the two-month peak near $4,450. Clearing $4,436 on a daily close opens $4,450, then $4,480 to $4,500, where the broken 200-day moving average and the psychological round number stack together. A sustained break above $4,485 extends the current phase and puts the January decline's retracement levels back in play.

Support is layered and tight. The $4,360 to $4,390 zone sits directly beneath current price and coincides with the near-term EMA structure — Tuesday's low at $4,386.10 tested the upper half of it and held. A clear break below $4,360 invalidates the bullish continuation setup.

Beneath that, the 2026 yearly open at $4,312 to $4,319 is the next reference, followed by the $4,244 to $4,224 support band. Loss of that zone reopens the six-week consolidation range at $4,070 to $4,125, where the $4,074 to $4,112 pivot held for a month and a half. The July low at $4,053.11 marks the floor of the entire structure.

Multi-timeframe technical ratings read buy on the one-week and one-month horizons, with hourly signals neutral and daily, weekly and monthly all reading strong buy. That configuration — bullish on longer frames, neutral intraday — describes an asset that has run and needs to digest.

Daily momentum has moved into oversold territory on some oscillator constructions and is beginning to reverse upward, which supports the case that Tuesday's 0.61% pullback is corrective inside the August advance rather than the start of a reversal.

The Polymarket contract on weekly XAU/USD range outcomes has traded $10,406 in volume.

The Fiscal Bid: $39 Trillion in Debt and $1 Trillion in Annual Interest

The structural argument underneath gold is arithmetic, and it operates independently of any inflation print or Fed decision.

US federal debt has surpassed $39 trillion and is approaching $40 trillion for the first time, with projections putting $50 trillion within reach shortly after. Annual interest servicing costs exceed $1 trillion. Those figures constrain the Federal Reserve's operating environment in ways individual data releases cannot alter — every basis point of increase at the long end raises the cost of rolling that stock.

The 30-year Treasury at 5.323%, the highest since 2007, is the market pricing that constraint. Long-dated sovereign yields rising across every major market simultaneously is the same repricing happening globally, and it is precisely the environment reserve managers diversify against.

The mechanism is straightforward. When a sovereign's debt trajectory forces either persistent inflation or persistent financial repression, the reserve asset that carries no issuer and no credit risk gains share. That is the explicit rationale behind Poland targeting 700 tonnes, behind South Korea's first purchases in thirteen years, behind Uzbekistan, Kazakhstan and the Czech National Bank adding to reserves, and behind the PBoC's 21-month streak.

Advanced-economy central banks were net sellers or neutral for decades. The dominant buying now comes from the Global South and from NATO members with direct exposure to geopolitical risk, and it is reshaping global reserve composition at roughly 1,000 tonnes per year against mine production of 3,000 to 3,500 tonnes.

The supply side offers no relief. Annual mine production has stagnated with limited new major discoveries. Recycling adds 1,000 to 1,500 tonnes. Central banks are absorbing between a quarter and a third of new supply on a structural basis.

This is why gold's floor keeps rising even through 27.7% drawdowns. The 2026 decline from $5,602.23 to $4,053.11 met sovereign accumulation the entire way down, and 89% of reserve managers expect to add further over the next twelve months.

That bid does not set the daily price. It sets the level below which the daily price does not stay.

Jackson Hole and the Minutes: Two Events That Set the Range

The next 72 hours carry more weight than the past two weeks of price action.

The July FOMC minutes release Wednesday, August 19. The market will read them for the size and conviction of the hawkish bloc. Cleveland Fed President Beth Hammack dissented in June in favor of raising and has publicly argued the central bank should act immediately, warning that the longer inflation stays above target the harder it becomes to reduce. If the minutes show that view held broader support than the vote implied, September hold odds fall from 69.9% and gold loses the sole input driving its August recovery.

The dovish read is already priced. Hike odds at roughly 35%, down from 50% a week ago, with year-end tightening no longer fully discounted, means the market has done most of the repricing it can do without an actual cut coming into view. The asymmetry favors disappointment.

Fed Chair Kevin Warsh speaks at the Jackson Hole symposium, and that address carries the larger risk. Warsh has reaffirmed commitment to bringing inflation down without signaling near-term increases. Against Brent at $91.76 and an SPR at 1982 lows, any acknowledgment that energy-driven inflation requires a policy response reverses the entire August trade.

The data calendar fills in around it. July industrial production lands Tuesday. The Philadelphia Fed Manufacturing Index for August, initial jobless claims, the Conference Board Leading Economic Index and preliminary August PMIs follow through the week. Each carries the capacity to shift September odds at the margin.

Beyond this week, the August CPI release on September 10 and the subsequent FOMC meeting are the decisive catalysts. Everything between now and then is positioning.

The geopolitical calendar runs in parallel and cannot be scheduled. The US-Iran interim agreement expired Monday with no replacement, Israel struck Lebanon over the weekend, new sanctions are in preparation, and the Strait of Hormuz remains the pressure point. Gold has priced escalation as a standing condition rather than an event.

Gold Price Forecast: $4,360 and $4,500 Are the Boundaries

The forecast reduces to two levels and the event risk that decides which one gives way.

Upside case. Gold at $4,389.83 must first reclaim and hold above $4,400, then break $4,436.15 on a daily close to invalidate the double rejection that has capped two sessions. Clearing that opens the two-month peak at $4,450 and the upper Bollinger band at $4,485. The decisive level is $4,500, where the broken 200-day moving average and the round number stack together — a daily close above it converts the August advance from a recovery into a trend reversal and puts the $4,645 area in play. Sustained dollar weakness below DXY 99.29 and dovish FOMC minutes are the required inputs.

Base case target for August: $4,485. Bullish target on a confirmed break: $4,500 to $4,550.

Downside case. Failure at $4,436 for a third time puts the $4,360 to $4,390 support zone under immediate test, and Tuesday's low at $4,386.10 sits inside it. A clear break below $4,360 removes the EMA structure supporting the recovery and exposes the 2026 yearly open at $4,312 to $4,319. Beneath that, $4,244 to $4,224 is the next shelf, and loss of that band reopens the six-week consolidation range at $4,070 to $4,125 with the July low at $4,053.11 as the floor.

Downside target on a hawkish FOMC surprise: $4,312 initially, $4,224 on continuation.

The structural floor is not technical. Central banks bought 289 tonnes in the second quarter and 533 tonnes across the first half. The PBoC has added for 21 consecutive months and took 20 tonnes in July alone. China imported 764 tonnes in six months against global mine supply of roughly 1,600 tonnes for the period. That bid strengthened as price fell from $5,602.23 to $4,053.11 and shows no sign of reversing.

Verdict: gold is a Fed trade in the short run and a fiscal trade in the long run. The August advance from $4,053.11 came almost entirely from September hike odds falling to 35%, and that repricing is largely complete. Wednesday's minutes and Warsh at Jackson Hole decide whether $4,500 or $4,312 prints first. The 30-year at 5.323% argues for the lower number. The sovereign bid argues the metal does not stay there.

That's TradingNEWS