Key figures
CB net buy 2025
0.0t
-21% vs 2024
CB net buy 2024
0.0t
Peak multi-year era
Gold ETFs 2025
+0.0t
2nd-strongest year
Avg price 2025
$0/oz
Q4 avg $4,135
Synthesis
Executive read
Central banks bought 863.3 tonnes in 2025 (World Gold Council / Metals Focus) — down from the exceptional 1,000t+ pace of 2022–24, but still far above the 2010–2021 average of ~473t. The bid is historically elevated and geographically wide, not a single-country story.
2025's demand structure flipped toward investment: global gold ETFs added ~801.2t; bar and coin hit a 12-year high (~1,374.1t); jewellery fabrication fell ~19% on price. That split — official + investment strong, jewellery weak — is the live regime into mid-2026.
China still sets the narrative (reported 2,306t stock) even when monthly disclosed adds are tiny. India holds ~880.5t at the RBI and is the world's swing retail market — jewellery elasticity is the main demand risk, not RBI fire-sales.
Primary sources: World Gold Council, Gold Demand Trends: Q4 and Full Year 2025 (29 Jan 2026); Metals Focus; RBI FX reports; bank research via public secondary coverage. Limited statistical extracts for review/commentary only.
Five seats
Sentiment from five places
Same metal, five agents. Officials buy for sovereignty; ETF allocators trade real rates; futures trade momentum; sell-side writes ranges; retail feels the sticker. Disagreeing seats are the signal.
Central banks (WGC survey)
89% / 45%
global ↑ / own ↑
Gold ETFs (fund flows)
−74t
June 2026 flows
CFTC managed money (COT)
~+115–117k
MM net contracts
Banks + Kitco analyst poll
38% / 23% / 38%
bull / bear / sideways
Kitco retail poll + Asia physical
Bias fading
retail near-term
Official sector
Bullish structuralCentral banks (WGC survey) · WGC CBGR Survey 2026 · published 16 Jun 2026 (fieldwork Feb–May)
- 89% of respondents expect global central-bank gold reserves to rise over the next 12 months (still near historic highs; 95% in the 2025 survey).
- Record 45% expect their own institution’s gold holdings to increase; only 1% expect a decrease.
- 74% see moderate or significantly lower US dollar share of global reserves over five years; gold’s share expected to rise.
- Crisis performance, diversification, inflation hedge, and geopolitical-risk hedge dominate “why hold gold.”
- Vaulting shift: more domestic storage + multi-location diversification (BoE still #1 vault at 57%, domestic 49%).
Board read · Price-insensitive multi-year bid. Softens near-term price but rarely reverses.
Source: World Gold Council, Central Bank Gold Reserves Survey 2026 (76 respondents)
Portfolio investors
Cooling tacticalGold ETFs (fund flows) · WGC Gold ETF Flows · June 2026 (8 Jul 2026 print)
- June: physically backed gold ETFs saw ~US$8.9bn outflows; holdings −74t to 4,047t; AUM ~US$526bn (−13% m/m on price + outflows).
- H1 2026 still net positive (~US$8bn inflows; holdings +18t) — Asia’s strongest H1 on record; North America the weak region.
- Morgan Stanley (via Kitco, Jun 2026): without a rebound in ETF demand, $5,200 H2 targets get hard.
- 2025 was the contrast case: +801t global ETF demand — 2026 is digesting that euphoria under higher real-rate odds.
Board read · Western funds are no longer the rocket fuel; Asia still carries H1. Near-term tape risk if outflows extend.
Source: World Gold Council ETF holdings & flows; secondary bank commentary
Speculative futures
Still net longCFTC managed money (COT) · CFTC COT · mid/late Jun–early Jul 2026 prints
- Managed money (hedge-fund style) remains net long gold futures/options in the ~115k–117k contract zone (MacroMicro / weekly COT reads).
- Non-commercial / speculative net longs also elevated on popular trackers (~180k–194k range in late Jun–early Jul prints — methodology differs from MM).
- Seeking Alpha COT note (30 Jun 2026): managed money re-engaged bullishly after a flat H1 positioning tape — still the group that can force short-term swings.
- Historical pattern: MM often drives the violent legs; official + ETF flows set the floor.
Board read · Crowded-long risk if real rates spike; not a short book. Watch week-to-week delta more than absolute level.
Source: CFTC Commitments of Traders (disaggregated + legacy); market trackers
Wall Street sell-side
Split near-termBanks + Kitco analyst poll · Kitco News Weekly Gold Survey · 10 Jul 2026 (13 analysts)
- Latest Kitco Wall Street poll: 38% bullish next week, 23% bearish, 38% sideways — indecisive after failed breakout / consolidation.
- Mid-June snapshot was much more bearish (70% of analysts expected a decline after hawkish Fed pricing).
- House targets still mostly constructive long-term: GS ~$4,900 YE (cut), JPM path toward ~$6,000 late-2026, UBS path-dependent mid-$5ks+.
- Metals Focus (via Kitco, Jul 2026): rangebound summer more likely than a clean bullish breakout while policy stays tight.
Board read · Structural bulls, tactical fences. Street is not all-in euphoria — good for avoiding top-call hysteria, bad for momentum chasers.
Source: Kitco News Weekly Gold Survey; bank research via public secondary coverage
Main Street / retail
Softened / price-sensitiveKitco retail poll + Asia physical · Kitco Main Street poll · 10 Jul 2026; WGC + Board Q1 India read
- Kitco online retail poll (10 Jul, n=282): Main Street renounced its usual bullish bias after another weak week — rare flip vs the classic “retail stubbornly long” pattern.
- Earlier (18 Jun): Main Street was still 54% bullish while Wall Street was 70% bearish — classic split; that gap has narrowed.
- India jewellery demand −16% in Q1 2026 on record local prices (Board / WGC desk); RBI stock high but retail is elastic, not official.
- China: investment channels (ETFs, bars) stayed more resilient than pure jewellery — Asia ETF H1 strength is the retail/semi-institutional cousin of that story.
Board read · High prices are rationing discretionary physical. Investment demand can still offset — but jewellery is a drag, not a tailwind.
Source: Kitco Main Street poll; WGC demand trends; RBI/India physical market reporting
Composite takeaway
Structural demand (seat 1) is still the floor. Tactical demand (seats 2–5) is no longer one-way bullish. Physical premiums below are the sixth check — whether stackers will actually pay up for metal.
Context
More signals worth watching
Reserve mix regime
WGC survey: most respondents see gold’s share of reserves rising over five years while the dollar’s share falls. That is a multi-year portfolio regime call, not a one-week trade.
Custody & sovereignty
India repatriated ~168t of gold to domestic vaults in FY25–26; WGC survey shows rising preference for domestic / diversified overseas storage. Custody location is becoming a sentiment signal of its own.
Opaque official demand
WGC estimated unreported official buying was still a large share of 2025’s central-bank total (~57%). Disclosed China/Poland tables understate the true official bid.
What would flip the board
Sentiment turns if: (1) ETF outflows run multi-quarter, (2) managed money nets go flat/short on a Fed scare, and (3) a major disclosed official seller appears. One of three is noise; three of three is regime change.
Internal links
Gold desk on The Board
Every page below is part of the gold / de-dollarization cluster and links back to this hub via the gold desk callout. Use this index for depth; use the sections above for the live desk read.
Official sector
Who is buying — country ledger
| Institution | Recent add | Holdings | Reserve share |
|---|---|---|---|
| China (PBOC) | +27t (2025 reported) | 2,306t | ~9% of total reserves |
| India (RBI + retail) | Stock ~flat YoY; +0.3t Q4’25→Q1’26 | 880.52t (Mar 2026) | Gold ~16.7% of FX reserves (value, Mar 2026) |
| Poland (NBP) | +102t (2025) | 550t | ~28% of reserves (target raised toward 30%, then 700t ambition) |
| Turkey (CBRT + Treasury) | +27t (2025, data through Oct) | ~644t official | Core reserve diversifier under lira/inflation stress |
| Kazakhstan | +57t (2025, record) | Rising (ministry cap enabled up to ~67t buy) | Net buyer stance tied to global tensions |
| Brazil | +43t (Sep–Nov 2025) | 172t | ~7% of total reserves |
| Azerbaijan (SOFAZ) | +38t (Q1–Q3 2025) | Rising (Q4 pending at WGC print) | SWF diversification, not pure CB |
| Czech Republic | +20t (2025) | 72t | Path to 100t by 2028 |
China (PBOC)
Strategic accumulator — slow public paceQ4’25 reported add was only ~3t — lowest quarterly public step since early 2024. Continues multi-year diversification; Chinese gold ETFs saw strong retail/institutional inflows even when PBOC pace cooled. Unreported official buying globally remains large (WGC).
Sources: WGC Gold Demand Trends FY2025; Metals Focus
India (RBI + retail)
Large reserve stock + price-sensitive retailRBI total holdings at a record ~880.5t. 2025–26 focus was repatriation (~168t brought home from BoE/BIS custody) more than net new buying. Retail jewellery demand fell ~16% in Q1 2026 on record local prices; bar/coin investment partially offset. RBI publicly denied mid-2026 ‘sold gold’ rumours — physical stock unchanged.
Sources: RBI half-yearly FX report; WGC/Trading Economics; Reuters (Jun 2026)
Poland (NBP)
Largest reported buyer 2024–2025Second consecutive year as top disclosed buyer. Governor Glapiński has framed further accumulation (toward 700t) as national-security policy — no hard deadline. Textbook EM ‘sovereign hard-asset’ buyer.
Sources: WGC Gold Demand Trends FY2025; NBP disclosures
Turkey (CBRT + Treasury)
Steady official buyer; volatile retail historyOfficial sector kept buying through 2025 volatility. Separately, Turkey’s retail/ETF episodes can dump physical faster than Western funds — see Board coverage of large short-window sales. Treat official vs private flows as different regimes.
Sources: WGC FY2025; Board markets desk
Kazakhstan
Domestic production + reserve buildHighest annual buying on WGC series back to 1993 after sales of domestic production gold were suspended. Governor: stay net buyer until tensions ease.
Sources: WGC FY2025; national bank statements via WGC
Brazil
Re-entry buyer (last bought 2021)Sizeable absolute add still leaves gold a small reserve share — room to keep buying without hitting Poland-style allocations.
Sources: WGC FY2025
Azerbaijan (SOFAZ)
Sovereign wealth fund (public gold book)Only major SWF with transparent gold disclosure in the WGC sample — useful signal that official-sector demand is broader than central banks alone.
Sources: WGC FY2025
Czech Republic
Program buyer to 2028 targetSteady multi-year schedule — less price-reactive than opportunistic EM buyers.
Sources: WGC FY2025; CNB
Structure
Stock vs flow
Western central banks and the IMF still hold the bulk of stocks. The 2022–2026 price regime has been driven by flows from EM official buyers, ETFs, and bar/coin.
- United States~8,133tLargest official stock; not an active net buyer
- Germany / IMF / Italy / FranceLarge legacy stocksWestern holders dominate stock, not 2022–25 flow
- China (reported)2,306tMarket consensus: true stock may be higher; disclosure is controlled
- India (RBI)880.5tRecord stock; 2026 story = custody location + retail elasticity
- Turkey (official)~644tHigh absolute stock relative to economy size
- Poland550tFast climber; allocation-driven
WGC: unreported official buying was ~57% of 2025's estimated central-bank total.
Market structure
Demand split
Official sector
863.3t net in 2025; Q4 alone 230.3t. WGC 2026 survey: 89% expect global official reserves to rise; record 45% plan own increases.
Investment
ETFs +801.2t in 2025; bar & coin 1,374.1t (12-year high). Safe-haven and diversification motives dominated.
Jewellery
Fabrication 1,638t (-19% y/y). Volumes cracked — India −16% jewellery demand Q1 2026.
Street research
Bank price targets
What institutions publish — not Board targets. Averages cut mid-2026 as Fed-cut hopes faded; many year-end bull cases stayed elevated.
| House | Target / path | As of | Thesis |
|---|---|---|---|
| Goldman Sachs | ~$4,900/oz YE 2026 | Jun 2026 (cut from $5,400) | Fewer/no 2026 Fed cuts; structural CB + ETF demand still the floor |
| J.P. Morgan | ~$6,000/oz by late 2026 (base path) | May–Jul 2026 revisions | Near-term demand soft; H2 re-acceleration via CBs + ETFs; long-term still bullish into 2027 |
| UBS | ~$5,500–$6,200 range (path-dependent) | H1 2026 updates | Still constructive; quarterly markers have moved with rates/ETF tape |
| Bank of America | Cut 2026 average (long-term upside retained) | Jul 2026 | Near-term cool-down after record run; structural story intact |
| ANZ / Wells Fargo (select) | Higher tails ($5,600–$6,300 scenarios) | H1 2026 | Bull cases keep $6k+ in play if investment demand re-ignites |
Board model
Scenarios — mid-2026
Working probabilities for communication, not a live prediction-market contract.
Base — elevated floor
~50–55%Consolidate under highs; institutional bid keeps deep drawdowns limited
CB net buying remains historically high (well above 2010–21 ~473t average) even if below 1,000t peak years; ETFs choppy not collapsing; jewellery stays weak on price.
Bull — second leg
~25–30%New highs if H2 ETF + EM retail investment re-accelerate
Geopolitical shock (energy chokepoint / major conflict premium), faster de-dollarization narratives, or clear Fed easing path. Bank bull cases cluster ~$5.5k–$6.3k.
Bear — bid fractures
~15–20%Sharp mean-reversion if a major official seller appears
Sustained PBOC pause that markets read as permanent; large reported official sales; or aggressive real-rate shock that kills ETF demand while jewellery is already weak.
Risk
Falsifiers & watchlist
- Sustained multi-quarter official-sector net selling (not one-off coin-minting sales)
- China reported reserves flat or falling for 2+ consecutive quarters without offsetting unreported estimates
- Global gold ETFs reverse 2025’s ~800t inflow into multi-quarter outflows
- India retail investment fails to offset jewellery weakness into major festival seasons
- Real US rates grind higher while risk assets stay calm — gold loses both hedge and FOMO bid
Library
Deep analysis on The Board
- Gold forecast 2026–2027: central bank record buying (Q1 data)
- Gold price path after the rally — 2026 update
- Gold price forecast next 5 years (2029–2031)
- Gold price forecast 2024–2029 framework
- Will BRICS replace the dollar? Reality-check scorecard
- BRICS de-dollarization analysis
- US dollar future: CBDC, gold standard, hyperinflation scenarios
- Turkey dumped more gold than every ETF combined
- If Hormuz closes: oil shock numbers (geo → metals)
FAQ
Common questions
What is the gold market prediction for 2026?
As of mid-2026, major banks still cluster year-end targets roughly in the mid-$4,000s to ~$6,000/oz depending on Fed path and ETF demand. Structural support is central-bank buying (863t in 2025 per WGC) plus investment demand; jewellery remains the weak leg at record prices.
Which central banks are buying gold?
In 2025 the largest reported buyers included Poland (+102t), Kazakhstan (+57t), Brazil (+43t), Azerbaijan's SOFAZ (+38t), Turkey (+27t), and China PBOC (+27t reported). Unreported official buying remains a large share of the global total.
How much gold does China hold?
Reported PBOC gold reserves stood at 2,306 tonnes at end-2025 (~9% of total reserves). Many analysts treat reported figures as a lower bound because disclosure is controlled.
What about India's gold reserves?
RBI holdings reached a record ~880.5 tonnes by March 2026. Recent policy emphasis has been repatriating gold from foreign custody into India, not large net new purchases. Retail jewellery demand fell sharply in Q1 2026 on high prices.
What is gold market sentiment right now?
As of mid-July 2026 it is split by seat: central banks remain structurally bullish (89% expect global official gold reserves to rise), gold ETFs cooled in June after a still-positive H1, futures managed money is still net long, Wall Street is mixed week-to-week, and retail/jewellery demand is price-sensitive.
What is a normal gold premium over spot at dealers?
It depends on product and payment. As of mid-July 2026, best ACH offers on 1 oz secondary bars and random-year American Gold Eagles were roughly +1.1% to +1.4% over spot; major brand resellers (APMEX, JM Bullion) often quote +4% to +6% ACH on the same Eagle. Credit cards add more. Kilo bars sit near the lowest % premiums. Always compare the same product and payment method.
Disclaimer: Educational analysis only. Not investment, trading, or financial advice. Gold prices, reserve data, and dealer premiums change; verify against primary sources (WGC, central banks, dealers) before acting.