Gold Price Forecast 2026–2051: A 25-Year Projection for Buying, Selling, Supply, Demand, Central Banks, and Recycling

#Scenario analysis for gold prices in 2030, 2040, 2050, and 2051—including real yields, reserve diversification, ETFs, bullion, jewellery, mine supply, recycling, technology, and the future of gold transactions

By All Metal Prices
Scenario model: First edition · Version 1.0
Scenario horizon: Calendar years 2026–2051 (annual scenarios)
First published: August 24, 2026
Last substantively reviewed: August 24, 2026
Next scheduled review: August 2027, or earlier after a material change in central-bank demand, real interest rates, reserve policy, mine supply, recycling, benchmark administration, or gold-market regulation

Scenario and financial-information notice: This static first-edition study presents conditional annual scenarios for calendar years 2026–2051. It is not a live quote, personalized recommendation, or invitation to buy or sell gold, a security, a futures contract, a mining share, jewellery, scrap, or a digital token. Twenty-five-year commodity scenarios are highly uncertain; the value of this paper lies in its assumptions, ranges, transaction analysis and update discipline—not false precision.

Source limitation: The cited baseline is date-bound and third-party source methodologies, revisions, licensing terms and benchmarks can change. This page does not reproduce or license those third-party values. Verify the original source, quote type, timestamp, material identity, assay, fees, taxes and counterparty terms before acting.

---

#The answer in one paragraph

The most defensible gold forecast through 2051 is a set of conditional price regimes, not one continuously compounding target. In inflation-adjusted 2026 U.S. dollars, the All Metal Prices managed-fragmentation scenario places gold’s annual-average price at $4,000–$5,900 per troy ounce in 2030, $3,900–$6,300 in 2035, $4,000–$6,900 in 2040, $4,200–$7,600 in 2045, and $4,600–$8,800 in 2051. A sustained restoration of monetary credibility—characterized by high positive real yields, lower geopolitical risk, less official-sector accumulation, strong currencies, and weak investor allocation—could produce a 2051 regime around $2,200–$4,000 in real 2026 dollars. Persistent monetary stress, larger gold allocations by central banks and private investors, fiscal dominance, reserve fragmentation, or repeated financial shocks could instead support a 2051 regime around $7,800–$16,000. Daily prices can move beyond every band. The central thesis is not simply “gold goes up.” It is that gold remains a very large above-ground stock whose price is set by the terms on which households, institutions, and official reserve managers are willing to hold that stock. Mine production matters, but allocation, liquidity, real yields, currency confidence, and the willingness to recycle matter more at the margin.

#Forecast table: gold prices in 2030, 2035, 2040, 2045, and 2051

The ranges below are possible annual averages in real 2026 U.S. dollars per troy ounce. They are not daily trading limits, statistical confidence intervals, or guaranteed floors and ceilings.

The managed-fragmentation range is the central analytical case, not a probability-weighted promise. The upside band is wide because monetary crises and abrupt portfolio reallocations are nonlinear. The downside band is also wide because a strong currency and high real-rate regime could coexist with jewelry demand, mine closures, recycling restraint, and bargain buying that limit the decline.

#What those ranges could mean in nominal dollars

Long-horizon price claims are often misleading because they mix future nominal dollars with today’s purchasing power. The table below converts the central real-price range using a constant 2.5% annual inflation assumption. This is an illustration, not an inflation forecast.

A nominal 2051 price of $8,500 would not mean the same thing as $8,500 in 2026. The real-price framework is therefore the primary forecast. The illustrative table also shows midpoint conversions at 2%, 2.5%, and 3% annual inflation.

#Key findings

1. Gold is a stock-allocation market before it is a mine-flow market. The World Gold Council estimates that about 220,700 tonnes existed above ground at the end of 2025, while annual mine production was about 3,672 tonnes. Most gold ever mined remains available in some form, so small changes in desired ownership can matter more than a small change in annual output. (World Gold Council 2026a)
2. The 2026 starting point is not normal. Gold entered the forecast window after 53 new all-time highs in 2025, record demand value, unusually strong ETF and bar-and-coin buying, elevated official-sector purchases, and another volatile repricing in 2026. (World Gold Council 2026b; World Gold Council 2026c)
3. Positive real yields do not make gold mechanically weak. The U.S. 10-year inflation-indexed Treasury yield was 2.35% on August 20, 2026, yet gold remained historically expensive. Real yields still matter, but geopolitical risk, reserve demand, dollar expectations, momentum, and portfolio concentration can overwhelm them for long periods. (Federal Reserve Bank of St. Louis 2026a)
4. Central banks are a structural demand source, but not an automatic one-way bid. Net purchases were 863 tonnes in 2025 and 289 tonnes in Q2 2026. A 2026 survey found that 89% of respondents expected global official holdings to rise over the next year and 45% expected their own institution to add. Yet purchases can slow at high prices, data are revised, and some institutions sell tactically. (World Gold Council 2026d; World Gold Council 2026e)
5. ETF and OTC positioning can move price faster than fabrication demand. In July 2026, global gold ETFs held 4,068 tonnes after $3 billion of monthly inflows, while the World Gold Council estimated average daily gold liquidity at $356 billion. The marginal repricing can therefore occur through financial flows without an immediate change in mine output. (World Gold Council 2026f)
6. Jewellery demand is both consumption and savings behavior. High prices reduce tonnage, encourage lighter products and exchange programs, and raise recycling. But in many markets jewellery remains a household balance-sheet asset, so lower prices can quickly revive demand.
7. Mine supply responds slowly. Q2 2026 mine production reached a second-quarter record, yet industry all-in sustaining costs had risen to a record $1,785 per ounce in Q1. Permitting, declining grades, infrastructure, energy, labor, royalties, water, politics, and project lead times constrain the response to high prices. (World Gold Council 2026g)
8. Recycling is the market’s fast physical buffer. Recycled supply can rise when prices, economic stress, or local-currency weakness motivate selling, but expectations of higher prices can delay sales. A displayed gold price is not the same as a scrap settlement after purity, stones, fees, refining loss, and dealer margin.
9. Buying access will continue to fragment by wrapper. Allocated bars, unallocated London claims, ETFs, futures, one-ounce cash-settled contracts, vault accounts, jewellery, and tokenized claims all transmit gold exposure differently. The future of buying gold is likely to be more continuous and fractional, but not risk-free.
10. Selling will become more provenance-sensitive. Bar identity, chain of custody, responsible-sourcing audit, sanctions screening, recycled-content claims, and digital records should increasingly influence marketability and settlement speed. (LBMA 2025a; OECD 2016)
11. The best long-term dashboard is multi-variable. Watch real yields, the dollar, global liquidity, central-bank net demand, ETF holdings, bar premiums, mine costs, recycling, jewellery affordability, fiscal stress, geopolitical fragmentation, and gold’s share of investable and official assets.
12. A good forecast must be allowed to fail visibly. This model preserves version 1.0, publishes its annual inputs, and defines triggers for revision. Historical ranges should not be silently rewritten after the fact.

---

Table of contents

#1. What this projection is—and what it is not

A useful 25-year gold forecast must separate facts, assumptions, and conditional conclusions. The facts include observable 2025 and 2026 prices, supply, demand, holdings, market structure, interest rates, and official-sector behavior. The assumptions describe how those conditions might evolve. The conclusions are the price regimes that could be consistent with each assumption set.

This paper does not claim that a deterministic equation can convert tonnes into a gold price. Gold’s annual physical flow is small relative to the quantity already held. A price can rise even when mine production is increasing, or fall even when mines are struggling, because the immediate issue is whether existing and prospective holders want to increase or reduce their allocation. Academic research on gold repeatedly finds that its behavior changes by period and purpose: hedge, safe haven, inflation protection, reserve asset, speculative asset, jewellery, and collateral are not interchangeable concepts. (Baur and Lucey 2010; Baur and McDermott 2010; O’Connor et al. 2015)

The forecast also does not assume that gold’s historical return since the end of Bretton Woods will repeat. The World Gold Council reports a long-run nominal return of roughly 9% a year in U.S. dollars since 1971, but that interval includes the removal of an official price constraint, high inflation, disinflation, globalization, financialization, multiple crises, and an enormous expansion in global wealth. Historical performance is context, not a compounding instruction. (World Gold Council 2026h)

The output is therefore three deliberately different regimes:

• Restored monetary credibility: high real returns on trusted financial assets, lower geopolitical risk, disciplined fiscal policy, restrained official gold demand, and less private allocation.
• Managed fragmentation: continued geopolitical and reserve diversification, uneven inflation, alternating rate cycles, moderate mine constraints, recurring but not permanent crises, and gold retaining a meaningful portfolio role.
• Monetary stress / reserve diversification: repeated fiscal and financial instability, lower confidence in major reserve assets, larger gold allocations, stronger bar-and-coin participation, and high willingness to pay for unencumbered or politically neutral reserves.

No probability weights are assigned. A user can reasonably believe that the central case is too bullish or too bearish and still use the model, because the assumptions and representative milestone tables are published.

Back to top ↑

#2. Why gold cannot be forecast like an ordinary commodity

#Almost all mined gold still exists

Oil is burned. Wheat is eaten. Copper is dispersed through infrastructure and products, though much can be recycled. Gold is different. It is chemically durable, valuable at high concentration, and worth recovering from products and scrap. The World Gold Council estimates that about 220,700 tonnes existed above ground at the end of 2025. Approximately 45% was in jewellery, 17% was held by central banks, 23% was in bars, coins, and ETFs, and the remainder was in other uses and less-visible OTC holdings. (World Gold Council 2026a)

Annual mine production of roughly 3,700 tonnes adds less than 2% to that stock. This stock-to-flow structure has two consequences.

First, scarcity cannot be inferred from mine production alone. A very large quantity exists, and some portion can be mobilized at the right price. Second, “available” gold is not the same as “willingly offered” gold. A wedding necklace, a central-bank reserve bar, an ETF holding, a coin collection, a refinery work-in-progress lot, and an unallocated London claim may all be part of the broad gold stock but have radically different reservation prices and legal forms.

#The price clears a balance-sheet market

Every newly mined ounce becomes someone’s asset or product. If jewellery and technology users do not absorb it, bars, coins, central banks, ETFs, OTC investors, dealers, exchanges, or inventories must. When the desired stock of gold rises faster than new supply, price adjusts upward to induce existing holders to sell or to reduce fabrication demand. When desired allocation falls, price adjusts downward until jewellery, retail investment, short covering, mine restraint, or reduced recycling restores balance.

This is why a flow deficit is not required for a bull market and a flow surplus is not sufficient for a bear market. Published gold supply-and-demand tables include OTC investment and stock changes because visible fabrication and disclosed investment channels do not explain the whole market. The model in this paper follows the same logic: jewellery, technology, central banks, and bars and coins are explicit; the remaining tonnes are labeled institutional/ETF/OTC absorption required. That residual is not an ETF forecast. It is the amount that must be held somewhere for the annual physical model to clear.

#Gold’s opportunity cost is financial

Gold does not pay a coupon or dividend. Its opportunity cost therefore rises when safe, liquid, inflation-adjusted yields rise—unless investors doubt the safety, liquidity, inflation adjustment, currency, or political neutrality of those alternatives. The relationship is powerful but not stable. In 2026, high positive U.S. real yields coexisted with historically high gold prices, illustrating that reserve demand, geopolitical risk, currency expectations, and momentum can offset the rate headwind.

The correct question is not “Are rates high?” It is “What real return is available on assets that the marginal gold buyer trusts, in the currency and jurisdiction that matter to that buyer?”

Back to top ↑

#3. The August 2026 starting point

#A market near $4,600 is already pricing substantial concern

The model date follows a two-year repricing. The LBMA PM annual average was $3,431 per ounce in 2025, 44% above 2024, after 53 new all-time highs. Gold then experienced a volatile 2026: the World Gold Council describes an early-year peak above $5,500 intraday, a retreat near $4,000 by late June, and continuing elevated volatility. By late August, the market was again around the mid-$4,000s. (World Gold Council 2026b; World Gold Council 2026c; LBMA 2026a)

That starting price matters. A forecast written at $1,200 could plausibly assume a large rerating before confronting affordability. A forecast written near $4,600 must ask how much monetary risk, diversification, and scarcity are already capitalized. The central scenario therefore does not begin by projecting another immediate doubling. It allows for consolidation, deep drawdowns, and long periods in which nominal gains are smaller than inflation.

#Demand is high in value, mixed in volume

Total gold demand including OTC exceeded 5,000 tonnes in 2025 for the first time, worth $555 billion. ETF holdings rose by 801 tonnes, bar-and-coin buying reached a 12-year high, and central-bank purchases were 863 tonnes. Jewellery tonnage fell, but spending reached a record $172 billion. (World Gold Council 2026b)

The first half of 2026 continued the tension between strong investment and weak affordability. Q2 demand including OTC was 1,269 tonnes; ETFs lost 45 tonnes during the quarter, bars and coins absorbed 307 tonnes, central banks bought 289 tonnes, jewellery consumption fell to 278 tonnes, and technology used about 80 tonnes. The average Q2 LBMA PM price was $4,506.29 per ounce. (World Gold Council 2026i)

This composition is a warning against interpreting “record demand” as one uniform phenomenon. The value of transactions can surge while physical jewellery tonnage falls. ETF holders can sell while central banks and Asian retail buyers add. A high price reallocates gold between users.

#Supply is responding—but slowly

Total supply was 5,002 tonnes in 2025: 3,672 tonnes of mine production, 1,404 tonnes of recycled gold, and a negative contribution from producer hedging. The recycling increase was only 3% despite a much larger rise in the dollar gold price, which suggests that expectations, household finances, local prices, and product attachment matter alongside the headline benchmark. (World Gold Council 2026j)

In Q2 2026, mine production rose 2% year over year to 966 tonnes, while recycling fell 6% to 326 tonnes. H1 mine production reached 1,867 tonnes. Industry all-in sustaining costs were $1,785 per ounce in Q1, 16% higher than a year earlier. (World Gold Council 2026g)

The forecast baseline rounds 2026 mine supply to 3,740 tonnes, recycling to 1,400 tonnes, and net de-hedging to 70 tonnes, yielding modeled total supply of 5,070 tonnes. These are annualized scenario inputs, not a claim that the final reported 2026 table will match exactly.

#Real yields remain a major headwind

The U.S. 10-year Treasury yield was 4.69% on August 20, 2026, and the 10-year inflation-indexed yield was 2.35%. (Federal Reserve Bank of St. Louis 2026a; Federal Reserve Bank of St. Louis 2026b) A gold price near record territory despite those yields implies that buyers are paying for more than near-term inflation. The embedded premium can include geopolitical insurance, reserve diversification, concern about debt trajectories, currency hedging, portfolio concentration, momentum, and the value of an asset without an issuer.

A downside scenario does not require deflation. It requires trusted real yields and currencies to remain compelling enough that gold’s insurance premium compresses.

#The market is exceptionally liquid, but liquidity is layered

The World Gold Council estimates that gold trading averaged roughly $361 billion a day in 2025. In July 2026, it estimated $356 billion a day across OTC, exchanges, and ETFs. LBMA’s 12-week moving average of reported gold trades for the period ending August 14, 2026 was $928.4 billion, while June London clearing transferred an average 16.7 million ounces a day worth $70.9 billion. These statistics measure different layers and should not be added as if they were one number. (World Gold Council 2026f; LBMA 2026b; LBMA 2026c)

Liquidity also depends on the instrument. A 400-ounce Good Delivery bar in a London vault, a one-ounce minted coin, an ETF share, a futures position, and a 14-karat chain do not have the same spread, settlement, or counterparty risk. The future of buying and selling gold will expand access while preserving these differences.

Back to top ↑

#4. Forecast methodology

#The model begins with holdings and annual absorption

The physical model has seven visible components:

1. Mine supply
2. Recycled supply
3. Net producer hedging or de-hedging
4. Jewellery demand
5. Technology demand
6. Central-bank net demand
7. Bar-and-coin net demand

A final residual—institutional, ETF, and OTC absorption required—balances total modeled supply. It can include ETF creation or redemption, OTC positions, exchange inventory, dealer stocks, institutional vaulted metal, and statistical differences. The residual should not be read as a forecast for any one product.

For each scenario, values are specified at six anchor years: 2026, 2030, 2035, 2040, 2045, and 2051. Annual values between anchors are shown in the scenario tables as linear interpolations. Real markets will not move smoothly; interpolation makes the stated assumptions inspectable.

#The central physical path

The table does not forecast a persistent “deficit” because gold does not disappear after use. It shows who must absorb annual supply. If private and official holders will not absorb the residual at the stated price, price falls until recycling, mine supply, fabrication, or investment behavior changes. If they want to absorb more than the available flow, price rises until existing holders release metal or fabrication demand falls.

#Price bands are assigned after the physical model

There is no fixed rule such as “500 tonnes of ETF inflow equals a 10% price rise.” Price response depends on starting positioning, inventories, leverage, liquidity, currency movements, volatility, and whether sellers are price-sensitive. The model therefore assigns broad price regimes after considering:

• desired gold allocation relative to global financial wealth;
• official reserve demand and custody policy;
• real yields and currency confidence;
• geopolitical and financial risk;
• mine costs, replacement, and project lead times;
• recycling response;
• jewellery affordability;
• ETF, OTC, and futures positioning;
• market structure and leverage;
• the starting valuation after the 2025–2026 repricing.

The World Gold Council’s Gold Long-Term Expected Return framework also treats long-run demand as broader than inflation alone, linking it to global nominal GDP and the scale of financial markets. The Council explicitly states that its framework is not a price forecast. This paper uses the same conceptual insight—global wealth and desired allocation matter—without adopting the Council’s model outputs. (World Gold Council 2024)

#Scenario assumptions

The restored-credibility scenario assumes:

• positive real yields remain attractive across major reserve currencies;
• fiscal risk premia decline rather than become embedded;
• geopolitical fragmentation eases;
• central-bank purchases normalize toward roughly 300 tonnes a year by the 2040s;
• private gold allocation falls or fails to keep pace with global financial wealth;
• lower real prices revive jewellery demand and restrain recycling;
• mine supply remains healthy before gradually declining.

The managed-fragmentation scenario assumes:

• alternating inflation and disinflation rather than one permanent regime;
• central-bank net buying remains structurally above pre-2022 norms but trends lower over time;
• gold remains around a few percent of private investable assets rather than becoming dominant;
• jewellery tonnage recovers gradually as incomes grow and products adapt;
• technology demand grows slowly;
• mine supply peaks during the 2030s and then faces replacement challenges;
• recycling rises steadily but does not flood the market;
• crises recur, but monetary institutions retain enough credibility to prevent continuous repricing.

The monetary-stress scenario assumes:

• repeated fiscal, currency, banking, or geopolitical shocks;
• a larger share of official reserves moves toward gold;
• private demand for bars, coins, ETFs, and allocated metal rises;
• high prices suppress jewellery tonnage and stimulate recycling;
• mine output initially rises, then declines as grades, costs, permitting, politics, and project scarcity constrain replacement;
• gold’s no-issuer status commands a durable premium.

#Why no probabilities are published

Assigning a 55% probability to the central scenario would imply more knowledge than the evidence supports. Over 25 years, elections, wars, reserve policy, technology, mine discoveries, capital controls, financial regulation, and institutional credibility can change repeatedly. The scenarios are better treated as stress tests. A reader can map current evidence to the assumptions and decide which regime is becoming more or less plausible.

---

Back to top ↑

#5. Real yields, currencies, liquidity, and macro risk

#Real yields are the clearest recurring variable—but not a complete model

Because gold produces no contractual income, higher real yields generally raise the cost of holding it. This relationship is strongest when the yield is available on assets perceived as safe, liquid, and politically reliable. It is weaker when inflation uncertainty is high, debt sustainability is questioned, currency risk matters, or investors are seeking insurance against the financial system itself.

The 2026 starting point demonstrates that distinction. A 2.35% U.S. 10-year real yield would historically be a substantial headwind, yet gold remained near $4,600. The market may be saying that the real yield is temporary, that fiscal or geopolitical risk deserves compensation, that reserve buyers are less rate-sensitive, or that global buyers face different local alternatives.

The downside scenario requires more than high nominal rates. It requires a sustained period in which inflation-adjusted returns are credible, currencies are trusted, and crises do not repeatedly reward gold holders.

#The U.S. dollar matters, but gold is not only a dollar trade

Gold is globally quoted in U.S. dollars, so a stronger dollar often weighs on the dollar gold price. But buyers in China, India, Turkey, Europe, the Gulf, Latin America, and Africa respond to local currencies, incomes, taxes, import rules, and political risk. A stable dollar price can coexist with a powerful local-currency bull market. Conversely, a weaker dollar can lift the benchmark while high local prices suppress physical demand elsewhere.

The IMF reported that the dollar’s share of disclosed foreign-exchange reserves rose to 57.13% in Q1 2026 from 56.42% in Q4 2025. It also cautioned that gold’s rise relative to U.S. Treasuries in official reserves during 2025 was driven largely by valuation effects. Gold is not included in the IMF’s COFER foreign-exchange reserve dataset, so claims about “de-dollarization” must distinguish transactions from price appreciation. (IMF 2026a; IMF 2026b)

The central scenario therefore assumes diversification without a sudden collapse in the dollar’s reserve role.

#Global growth supports both alternatives and demand

The IMF projected global growth of 3.0% in 2026 and 3.4% in 2027. Over longer periods, economic growth can support gold jewellery, technology, savings, and institutional portfolios, even while it improves the attractiveness of equities and credit. (IMF 2026c)

This dual effect helps explain why gold is not merely a recession trade. In severe stress, investment demand can dominate. In expansion, consumer and wealth growth can support demand. The composition changes.

#Volatility is part of the forecast, not an error around it

The World Gold Council notes that gold has experienced multiple drawdowns greater than 20% after record highs since 1971. A 2051 central range above the 2026 price does not imply a smooth path. Gold could fall 30%, remain below a prior high for years, and still finish within the central long-term regime. (World Gold Council 2026c)

Buyers and sellers should therefore distinguish a long-term thesis from timing. A correct regime forecast can coexist with a poor transaction price.

Back to top ↑

#6. Central banks and reserve diversification

Model relationship (August 24, 2026): The later, dedicated central-bank Gold reserves forecast is the authoritative All Metal Prices model for official-holdings paths. Its front-loaded net-buying path supersedes the central-bank-demand row in this first-edition price-balance model for holdings questions. The price bands and other demand components here have not been re-estimated, so do not combine the two model tables; the original row remains visible to keep this price model auditable.

Historical context: For the documented 1609–2026 development of monetary reserves, convertibility, sales and custody, read the history of central banks and precious metals.

#Official demand has changed the marginal market

Central banks bought more than 1,000 tonnes a year in 2022–2024 according to World Gold Council estimates, followed by 863 tonnes in 2025. That was far above the 400–500-tonne average of much of the preceding decade. Q2 2026 purchases rebounded to 289 tonnes after a weak and subsequently revised Q1. (World Gold Council 2026b; World Gold Council 2026e)

Official buyers differ from tactical funds. They may care about reserve diversification, sanctions resilience, collateral, crisis performance, and assets without another government’s liability. They can also buy through domestic production or off-market channels, making the true timing harder to observe.

#Survey intent is supportive, not binding

The 2026 Central Bank Gold Reserves Survey collected 76 responses. Eighty-nine percent expected global official gold holdings to rise during the next 12 months; 45% expected their own institution to add; 84% expected gold to represent a larger share of total reserves in five years; and 74% expected the dollar’s global reserve share to be lower. (World Gold Council 2026 detailed survey results)

Those responses are powerful evidence of strategic interest. They are not purchase orders. Reserve managers face price, liquidity, governance, custody, political, and portfolio constraints. A high gold price can increase gold’s reserve share without one tonne being purchased. The forecast accordingly treats central-bank demand as a range and declines it gradually in the central case rather than extrapolating 1,000 tonnes forever.

#Custody may become as important as quantity

The survey also reported more interest in domestic storage and diversified overseas vaulting. Over 25 years, reserve policy may focus increasingly on:

• location and legal title;
• ability to mobilize gold rapidly;
• swap and collateral eligibility;
• sanctions and counterparty exposure;
• bar specification and assay;
• domestic refining capacity;
• transparency versus strategic confidentiality.

A central bank can change the usefulness of its gold without changing the number of tonnes. This is one reason the future of selling and lending official gold cannot be inferred from headline holdings alone.

#What would make official demand reverse?

The restored-credibility scenario assumes that reserve managers find better combinations of safety, liquidity, and return in sovereign debt and diversified currencies. It also assumes geopolitical tensions ease and sanctions risk becomes less salient. Net official demand falls toward 300 tonnes a year by the 2040s—not necessarily because central banks abandon gold, but because they reach target allocations.

The stress scenario assumes the opposite: target allocations rise, more institutions view gold as an active rather than legacy reserve, and domestic or politically neutral custody becomes strategically valuable. In that scenario, annual central-bank demand remains 850–1,050 tonnes across much of the forecast.

Back to top ↑

#7. ETFs, OTC markets, futures, and institutional allocation

#ETFs made gold easier to own—and easier to sell

Physically backed ETFs separate economic exposure from personal custody. They allow rapid portfolio rebalancing, but they also permit large redemptions without the delays associated with selling coins or jewellery. Global ETF holdings rose by 801 tonnes in 2025, then experienced mixed flows in 2026. In July, holdings increased by 23 tonnes to 4,068 tonnes, still below the February record of 4,176 tonnes. (World Gold Council 2026f)

Over 25 years, ETF structures will probably remain important, but the wrapper may evolve. Fees, tax treatment, authorized-participant concentration, lending policy, vault location, audit, and redemption rules will continue to distinguish products that all appear to “track gold.”

#OTC London remains the core wholesale market

The LBMA Gold Price is set twice daily in auctions beginning at 10:30 and 15:00 London time and represents unallocated gold delivered in London. Most wholesale Loco London trading settles through unallocated accounts, where a client has a claim on a provider rather than title to specific numbered bars. Allocated accounts identify specific bars. (LBMA 2026a; LBMA 2026d)

That distinction will remain central. An unallocated balance may be liquid and operationally efficient, while an allocated holding reduces some credit exposure and creates different custody costs. Neither is automatically “better”; the right form depends on purpose.

#Futures are becoming more granular and continuous

The standard COMEX Gold futures contract represents 100 troy ounces. Micro Gold represents 10 ounces. CME’s one-ounce cash-settled contract became available for 24/7 trading in July 2026, giving smaller traders a more precise leveraged instrument. (CME Group 2026a; CME Group 2026b; CME Group 2026c)

This expands access, but it does not turn a cash-settled contract into a coin. Margin, leverage, liquidation, roll, basis, exchange rules, broker credit, and tax treatment remain material. The future of buying gold will likely include more precise exposure with less capital, which increases both accessibility and the speed of forced selling.

#Institutional allocation is the swing factor

The World Gold Council estimates that investor bullion—bars, coins, ETFs, and OTC holdings—represented about 3% of roughly $320 trillion in global financial assets at the end of 2025. (World Gold Council 2026k) A shift from 3% to 3.5% would be large relative to annual mine production. A shift back toward 2% could release a similarly meaningful quantity.

This is the core of the upside and downside cases. The long-term gold price is less about whether one mine opens and more about whether the world’s public and private balance sheets assign a larger or smaller role to gold.

---

Back to top ↑

#8. Jewellery, bars, coins, and household saving

#Jewellery is not merely decorative consumption

In many markets, jewellery combines adornment, social meaning, inherited wealth, emergency liquidity, and savings. That makes demand more resilient in value than in weight. Q2 2026 jewellery consumption fell to 278 tonnes—the lowest quarterly volume since the pandemic—while spending rose 14% year over year to $40 billion. (World Gold Council 2026i)

At high prices, households respond by buying lighter pieces, lower-carat products, old-gold exchange, smaller gifts, or investment bars instead of heavy jewellery. Fabricators reduce inventory and tighten working capital. At lower prices, deferred purchases can return. This demand elasticity creates a stabilizer that is absent from a pure financial model.

The scenario paths reflect that behavior. Jewellery demand falls to about 800 tonnes by 2051 in the monetary-stress case because real prices are very high. It rises to about 2,500 tonnes in the restored-credibility case because real prices are lower and global incomes are larger. Neither path assumes jewellery disappears.

#Bars and coins are both savings instruments and crisis products

Retail bullion demand responds to inflation, distrust, local banking conditions, capital controls, tax, dealer inventory, and price momentum. In some episodes, a rising price attracts buyers; in others, it discourages them. Premiums can diverge sharply from the London benchmark when mint capacity, logistics, or local regulation becomes the binding constraint.

A retail buyer does not pay “spot.” The acquisition cost can include fabrication, wholesale premium, dealer margin, shipping, insurance, payment cost, tax, and a local scarcity premium. The seller receives bid value after authentication, liquidity, and dealer margin. A round trip can therefore lose money even when the benchmark is unchanged.

The central scenario holds bar-and-coin absorption near 1,300–1,400 tonnes a year over much of the period. That is not a prediction of identical regional behavior. It represents a persistent global role for physical savings despite wider access to ETFs and digital products.

#Affordability changes product design

If gold remains expensive in real terms, the retail market is likely to adapt through:

• smaller bar and coin weights;
• fractional units and digital accumulation plans;
• lighter jewellery and hollow construction;
• higher use of exchange and buyback programs;
• clearer disclosure of melt value versus workmanship;
• products designed for automated authentication;
• more explicit separation between collectible premium and metal value.

This adaptation supports transaction volume without requiring the same tonnage per buyer.

Back to top ↑

#9. Mine supply through 2051

#High prices do not create mines quickly

Gold projects can require years of exploration, drilling, feasibility work, finance, permitting, infrastructure, community agreement, construction, and ramp-up. A price spike improves project economics but does not shorten every bottleneck. The World Bank’s 2026 outlook likewise described mine output as broadly stable despite high prices and expected supply to respond gradually. (World Bank 2026)

The 2026 baseline is favorable in one sense: mine output is high and margins above average. It is difficult in another: industry costs are also at records, and many existing operations face lower grades, deeper pits, longer haulage, water constraints, energy costs, taxes, social-license disputes, or political risk.

The U.S. Geological Survey estimated world gold reserves near 66,000 tonnes in its 2026 commodity summary, while the World Gold Council’s broader resource estimate was substantially larger. “Reserves” are not a fixed geological total; they are the economically recoverable portion under current knowledge, prices, technology, and assumptions. (U.S. Geological Survey 2026; World Gold Council 2026a)

#The mine-supply paths

In the restored-credibility scenario, mine production rises to 3,900 tonnes in 2030, then declines to 3,100 tonnes by 2051 as lower real prices reduce project incentives. In the central scenario, higher prices support output near 4,000 tonnes through 2040, followed by a decline to 3,600 tonnes as replacement becomes more difficult. In the stress scenario, output reaches 3,950 tonnes in 2035 but falls to 3,100 tonnes by 2051 despite high prices because geological, political, and permitting constraints dominate.

These paths make a critical point: an extremely high price does not imply ever-rising mine supply. It can instead signal that the market is bidding for a limited flow from a deteriorating or slow-growing asset base.

#What could lift mine supply above the model

The model would be too low if exploration success improves, large deposits are permitted faster, autonomous mining sharply lowers costs, ore sorting and recovery improve, or currently marginal resources become economic. Higher formalization of artisanal and small-scale mining could also increase recorded supply.

#What could push mine supply below the model

The model would be too high if grades decline faster, energy and capital costs remain elevated, resource nationalism increases, sanctions restrict equipment and finance, water becomes more constrained, community opposition delays projects, or environmental liabilities raise closure and permitting costs. A rising benchmark does not guarantee that producers capture the full increase; royalties, taxes, currencies, labor, and inputs can absorb much of it.

#Producer hedging remains a secondary but useful signal

Producers sometimes sell future output forward to secure financing or protect project economics. Hedging adds apparent supply during build-up and removes supply when positions are closed or delivered. The model includes only modest net hedging because the long-run direction is unpredictable and small relative to mine and recycling supply. A major return to project hedging would indicate that producers view forward prices as unusually attractive or lenders are demanding protection.

Back to top ↑

#10. Recycling and the future of selling scrap gold

#Recycled gold is the quickest physical response

Recycling represented about 28% of 2025 total supply. Unlike a mine, a household or business can sell existing gold quickly. But the decision depends on more than price. Economic distress, local currency, taxes, sentimental value, replacement cost, expectations, dealer access, and trust all matter.

The muted 2025 recycling response to a 67% increase in the dollar price is instructive. Some holders expected further appreciation; others did not need liquidity; and some markets faced lower local incentives. In Q2 2026, recycling fell year over year as the price declined from its earlier peak. (World Gold Council 2026j; World Gold Council 2026g)

#How a scrap-gold offer is calculated

A transparent offer can be expressed as:

Gross contained value = gross weight × purity × benchmark price per unit

Expected settlement = gross contained value × payable percentage − fees and deductions

For a 100-gram lot of 18-karat jewellery:

• nominal purity: 75%;
• fine-gold content before adjustment: 75 grams;
• troy ounces: 75 ÷ 31.1034768 = approximately 2.411 ounces;
• at a hypothetical benchmark of $4,600 per ounce: gross contained value ≈ $11,091;
• at a 94% payable after assay, refining, and dealer margin: approximately $10,426 before any stone-removal or service charge.

The example is not a live quote. Real settlements depend on actual assay, stones and non-gold parts, payment timing, lot size, market, tax, and counterparty.

#Future sellers will need better evidence

By 2051, professional scrap transactions are likely to rely more on:

• calibrated X-ray fluorescence screening;
• fire assay or other representative laboratory methods for larger lots;
• digital scales and immutable weight records;
• photographs and item-level chain of custody;
• sanctions, theft, and anti-money-laundering screening;
• automated melt-value calculations;
• clearer payable percentages and fee disclosure;
• documented destruction of branded or sensitive components;
• provenance and recycled-content certificates.

These systems can reduce fraud and speed settlement, but they do not eliminate sampling risk. A heterogeneous lot can produce an accurate test on an unrepresentative sample.

#Jewellery, coins, and collectibles should not be sold as generic scrap automatically

The melt price is only one possible value. A signed jewellery piece, rare coin, historically important bar, or collectible medal may command value above metal content. Conversely, a widely available modern item may sell below retail replacement value even when the gold price is high.

A rational seller identifies the product before choosing the market:

1. collectible or historical sale;
2. resale as wearable jewellery;
3. dealer bullion bid;
4. refinery or scrap settlement.

The highest benchmark does not compensate for choosing the wrong channel.

Back to top ↑

#11. Technology demand and urban mining

#Technology is small in tonnage but strategically durable

Gold’s conductivity, corrosion resistance, malleability, and reliability support electronics, connectors, semiconductor packaging, medical applications, aerospace, and other specialized uses. Q2 2026 technology demand was about 80 tonnes, with AI-related demand offsetting weakness in consumer electronics. (World Gold Council 2026i)

The central scenario increases annual technology demand from 322 tonnes in 2026 to 600 tonnes in 2051. That path assumes growth in electronics and high-reliability systems, partly offset by thrifting and substitution. It is deliberately modest relative to total gold demand.

#AI can increase demand without making gold an industrial metal

Data centers, high-performance computing, advanced packaging, and network equipment may use more gold-bearing components. Yet manufacturers continuously reduce precious-metal loadings when costs rise. Even a doubling of technology demand would remain smaller than shifts in central-bank, ETF, or jewellery demand. Claims that AI alone will determine gold’s price should therefore be treated skeptically.

#E-waste is a growing secondary resource

UNITAR and the International Telecommunication Union estimated that 62 million tonnes of e-waste were generated in 2022 and projected 82 million tonnes by 2030. They estimated $15 billion of gold embedded in 2022 e-waste, while only 22.3% of the total mass was documented as properly collected and recycled. (UNITAR and ITU 2024)

That does not mean all embedded gold is economically recoverable. Collection, concentration, hazardous materials, labor, energy, regulation, and smelter terms determine recovery. Over the forecast period, better sorting and formal recycling should raise recovery, but miniaturization and thrifting can reduce gold per device.

#Urban mining will become more integrated

Future gold sellers are likely to include electronics manufacturers, data-center operators, recyclers, telecommunications firms, and equipment lessors. Closed-loop contracts can return recovered gold to the original manufacturer, reducing price exposure and provenance uncertainty. This market will be less visible than retail scrap but increasingly important for responsible supply.

Back to top ↑

#12. Responsible sourcing, provenance, and bar identity

#Provenance is moving from policy document to transaction field

The OECD’s minerals guidance sets a risk-based due-diligence framework for companies sourcing from conflict-affected and high-risk areas. LBMA Good Delivery refiners are subject to a responsible-sourcing program with independent audit and public reporting. (OECD 2016; LBMA 2026e)

Over 25 years, source attributes are likely to become more machine-readable. Buyers may increasingly ask not merely “Is this .9999 fine?” but:

• Which mine, recycler, or aggregator supplied it?
• Which refinery processed it?
• Was it mined, recycled, or mixed?
• What audit and chain-of-custody evidence exists?
• Is the origin restricted by sanctions or internal policy?
• What environmental or human-rights claims can be substantiated?

#Gold Bar Integrity creates a foundation

LBMA and the World Gold Council launched the Gold Bar Integrity initiative to create records of bar origin and chain of custody. LBMA’s GBI Database launched in January 2025 and was described by LBMA as the beginning of a broader move toward a more transparent and efficient market. (LBMA and World Gold Council 2022; LBMA 2025a)

A digital record does not prove the truth of every upstream claim. It improves traceability only if identities, audits, and data inputs are reliable. The long-term market will need governance for data correction, privacy, competing standards, and bars that predate digital systems.

#Artisanal mining remains the hardest inclusion problem

The Minamata Convention identifies artisanal and small-scale gold mining as the largest source of mercury emissions and releases and requires affected parties to develop national action plans. (Minamata Convention 2026) Excluding all artisanal gold from formal supply chains can push vulnerable miners toward opaque buyers; accepting it without credible due diligence can perpetuate harm. The future market needs pathways for formalization, mercury reduction, legal title, fair payment, and auditable aggregation.

#Provenance could create differential liquidity

Gold is chemically fungible after refining, but bars need not be commercially identical. A bar with accepted refiner status, complete custody history, current sanctions screening, and verified ownership may settle faster and at tighter spreads than an orphaned or disputed bar. The likely result is not a universal “green premium,” but a hierarchy of marketability and compliance cost.

Back to top ↑

#13. How buying gold may change

#Physical bullion: smaller, authenticated, and easier to finance

Retail bars and coins will remain because direct possession is the point for many buyers. Product development is likely to emphasize smaller weights, tamper evidence, machine-readable identifiers, buyback guarantees, and digital ownership records. The risk is that convenience fees and fractional fabrication premiums rise as a percentage of metal value.

Large buyers will continue to prefer wholesale bars and vault accounts. Their concerns will include title, location, insurance, audit, lending, and ability to move between markets.

#Allocated and unallocated accounts: clearer risk disclosure

Unallocated gold offers liquidity and operational simplicity but creates a credit relationship. Allocated gold assigns specific bars but adds custody and transfer complexity. Future regulation and institutional policy may require clearer disclosure of rehypothecation, sub-custody, bar lists, insurance, and bankruptcy treatment.

The buying decision should be based on purpose:

• trading liquidity;
• reserve custody;
• collateral;
• emergency access;
• long-term wealth storage;
• geographic diversification.

#ETFs: lower friction, persistent wrapper risk

ETFs will likely become cheaper and more specialized by vault location, currency hedge, environmental policy, or redemption design. They will remain securities, not personal possession. Buyers must distinguish the metal’s market risk from fund, custody, regulatory, tax, and trading risks.

#Futures and options: broader participation, faster leverage cycles

CME’s progression from 100-ounce to 10-ounce and one-ounce contracts shows how market access can become more granular. Twenty-four-hour trading may reduce weekend gaps in one product while increasing continuous risk management and automated liquidation. Smaller notional size does not remove leverage.

#Digital and tokenized gold: useful only when the legal chain is real

Tokenization can allow fractional transfers, programmable settlement, and integration with payment or collateral systems. The critical questions are old ones in new language:

• Who owns the underlying bars?
• Are they allocated?
• Who is the custodian and sub-custodian?
• Can holders redeem, in what size, and in which jurisdiction?
• Can tokens be frozen?
• Is the supply independently reconciled?
• What happens in insolvency?

A token can improve the transaction layer while weakening the ownership layer. “Backed by gold” is not a complete legal description.

#Jewellery buying: more transparent separation of metal and craftsmanship

Digital pricing tools will make melt value easier to calculate at purchase. High-quality sellers may show metal weight, fineness, workmanship, tax, brand premium, and buyback terms separately. This could improve consumer understanding but also reveal how little of a retail price is recoverable as scrap.

Back to top ↑

#14. How selling gold may change

#Bullion sellers will expect continuous bids

With more 24/7 instruments, retail platforms may quote indicative bids continuously. Physical settlement will still be constrained by business hours, verification, bank transfers, logistics, and local law. The spread can widen sharply when the reference market is moving but the dealer cannot hedge or replenish inventory.

#Jewellery sellers will receive faster but more differentiated offers

Automated testing and pricing can shorten the transaction. Offers may be segmented into:

• immediate cash or instant transfer;
• higher delayed refinery settlement;
• consignment as jewellery;
• auction or branded resale;
• exchange credit toward a new purchase.

The fastest offer will not always be the highest. Sellers should compare net proceeds, time, recourse, data use, and custody risk.

#Industrial sellers will sell data with the metal

For e-waste, sweeps, plating residues, dental alloys, and manufacturing scrap, the commercial product is a documented lot. Weight history, process source, assay protocol, sampling method, moisture, deleterious elements, and chain of custody can determine the payable value. More transparent settlement portals should reduce information asymmetry, but sophisticated sellers will still audit sampling and assay.

#Miners will sell a portfolio of outputs and risks

Mining companies can sell doré, concentrate, by-products, royalties, streams, and future production. They may hedge currencies, energy, or gold. Over time, carbon, water, biodiversity, community, and provenance data may influence offtake and finance. A high benchmark does not eliminate the cost of capital.

#Central banks and institutions will sell less often but with greater market impact

Official sales can be tactical rebalancing, currency intervention, reserve management, or policy signals. Because the official sector is a large holder, a shift from net buying to net selling would be one of the most important bearish indicators in the model. Transparent coordinated selling agreements are possible, but strategic opacity is also likely.

Back to top ↑

#15. Scenario A: restored monetary credibility

#Price path

This is not a forecast that gold returns to its pre-2020 nominal price. It is a forecast that the real insurance and allocation premium compresses. At 2.5% inflation, the 2051 real range converts to roughly $4,080–$7,415 nominal.

#Conditions required

The scenario requires several developments to persist together:

• credible fiscal consolidation or debt stabilization;
• high real yields without repeated financial accidents;
• low and stable inflation expectations;
• strong and trusted reserve currencies;
• reduced sanctions and geopolitical fragmentation;
• central-bank gold purchases falling toward 300 tonnes a year;
• private investors reducing gold’s share of portfolios;
• low crisis frequency and strong competing asset returns.

One or two of these conditions would not be enough. Gold can perform poorly during isolated periods of rising real yields, but a 25-year low-real-price regime needs a broad restoration of confidence.

#Physical market response

Lower prices revive jewellery to 2,500 tonnes by 2051 and restrain recycling to 1,750 tonnes. Mine production initially rises, then falls to 3,100 tonnes as project economics weaken. Institutional absorption drops to about 575 tonnes by 2051. This is still positive because new mine supply must be held somewhere; it is simply much smaller relative to global wealth.

#Buying and selling behavior

Buyers become more price-sensitive and focus on jewellery, collecting, or tactical diversification rather than urgent insurance. Dealers carry lower inventory values and may face consolidation. Scrap sellers receive lower real bids, so economically motivated recycling slows. High-premium fractional products become less attractive unless fabrication costs also fall.

#Why the downside is not lower

The range recognizes that lower real prices stimulate demand and discourage supply. Gold also retains monetary history, liquidity, and no-issuer characteristics. A permanent collapse toward mining cost is possible in theory but would require an extraordinary decline in desired holdings across jewellery, private investment, and official reserves.

Back to top ↑

#16. Scenario B: managed fragmentation

#Price path

This central case allows a broad real-price plateau in the early 2030s, followed by gradual upward pressure as global wealth grows, official demand remains meaningful, and mine replacement becomes harder. It does not assume that every decade is a crisis.

#Conditions

• real yields alternate rather than remain permanently high or negative;
• the U.S. dollar remains the leading reserve currency but its dominance erodes gradually;
• central banks buy 550–750 tonnes a year through much of the period;
• private gold allocation remains near current broad proportions;
• periodic crises sustain insurance demand;
• jewellery adapts to price rather than collapses;
• mine supply peaks around the 2030s;
• recycling rises as the above-ground stock and digital selling access grow.

#Physical market response

Total modeled annual supply rises from 5,070 tonnes in 2026 to about 5,700 tonnes in 2040, then declines to 5,500 tonnes in 2051. Jewellery and technology absorb more tonnes over time. Institutional/ETF/OTC absorption remains substantial but trends down from its peak as other demand grows.

#Buying and selling behavior

Gold becomes easier to own fractionally, but serious holders pay more attention to custody, title, and jurisdiction. Physical dealers coexist with ETFs, futures, vault platforms, and tokenized products. Recycling becomes professionalized, and provenance records reduce settlement friction for compliant metal. The benchmark remains global while retail spreads remain local.

#The most important feature: long plateaus are possible

A central long-term outcome need not generate exciting annual returns from a 2026 purchase near $4,600. The real midpoint rises from roughly $4,550 in 2026 to $4,950 in 2030, $5,450 in 2040, and $6,700 in 2051. That is about 1.6% annualized real growth over 25 years before ownership costs, tax, spreads, or financing. Gold can fulfill a reserve or diversification role without outperforming every productive asset.

Back to top ↑

#17. Scenario C: monetary stress and reserve diversification

#Price path

This is a regime of repeated repricing, not necessarily hyperinflation. Gold can reach a much higher real range if major holders decide that it deserves a larger share of reserves and portfolios.

#Conditions

• fiscal dominance or repeated debt-market stress;
• volatile inflation and declining confidence in real sovereign returns;
• broader use of sanctions and reserve immobilization;
• reserve diversification across central banks;
• rising domestic custody and strategic stockbuilding;
• persistent private bar, coin, ETF, and allocated demand;
• political or geological constraints on mine replacement;
• high willingness to pay for unencumbered gold.

#Physical market response

High prices suppress jewellery to 800 tonnes by 2051 and lift recycling to 2,700 tonnes. Central banks absorb 850–1,050 tonnes a year. Bar-and-coin demand reaches 2,200 tonnes by 2051. Mine production ultimately falls to 3,100 tonnes because high prices cannot instantly solve project scarcity, social license, or geology.

#Why the range is so wide

At high prices, behavior becomes nonlinear. Recycling can surge, governments can change taxes or import rules, central banks can slow purchases, and private investors can take profits. Conversely, a loss of confidence can produce a portfolio shift much larger than annual mine supply. A narrow $12,000 target would hide more than it reveals.

#Risks inside the bullish case

A stress-driven bull market can produce severe drawdowns. A crisis may initially force investors to sell liquid gold to meet margin calls. Governments can restrict trade, tax gains, requisition local production, or impose capital controls. Digital claims can be frozen. Counterfeit and fraud risk rises when public demand surges. “Bullish gold” is not synonymous with easy profits or easy settlement.

Back to top ↑

#18. Gold price forecast for 2030

#Central range: $4,000–$5,900 real 2026 dollars

The 2030 outcome is the most constrained by current evidence. Mine projects already in development, existing central-bank policy, known ETF infrastructure, and current rate cycles provide more visibility than the 2040s.

The central range allows the extraordinary 2025–2026 repricing to consolidate. A price near the lower end would imply that real yields remain high, geopolitical risk falls, and official buying moderates. A price near the upper end would imply that central-bank demand remains strong, ETF participation broadens, and the market treats the 2026 correction as a pause rather than a reversal.

The World Bank projected gold to rise about 37% on average in 2026 and decline 9% in 2027, while an LBMA survey of 16 professional analysts produced an average year-end 2026 forecast around $4,500, with a range of $3,879 to $5,100. Those short-term forecasts illustrate how quickly views can diverge even before 2030. (World Bank 2026; LBMA 2026f)

#What to monitor before 2030

• whether U.S. and global real yields remain positive;
• whether gold ETF holdings exceed the 2026 record sustainably;
• whether central-bank demand remains above 700 tonnes a year;
• whether China and India retail demand shifts from jewellery toward investment;
• whether recycling finally responds strongly to the higher price plateau;
• whether mine margins translate into project approvals;
• whether geopolitical risk falls or becomes structural.

Back to top ↑

#19. Gold price forecast for 2035

#Central range: $3,900–$6,300 real 2026 dollars

By 2035, the question shifts from cyclical repricing to allocation. Has gold become a larger strategic reserve, or has it merely experienced one exceptional decade? Have private investors maintained a 3% allocation as global wealth grows? Has mine supply expanded enough to reduce scarcity concerns?

The central case allows a real range slightly below or above 2030 because a long plateau is plausible. A return to $3,900 would not invalidate gold’s reserve role; it would imply that the initial premium was large and alternatives improved. A move toward $6,300 would imply continued absorption without a full monetary-stress regime.

#Structural signpost

The most important 2035 signpost is gold’s share of official and private assets, measured in both tonnes and value. A rising value share caused only by price is less powerful than broad net buying. Transaction data, reserve disclosures, ETF tonnes, and bar-and-coin tonnage should be examined alongside market value.

Back to top ↑

#20. Gold price forecast for 2040

#Central range: $4,000–$6,900 real 2026 dollars

By 2040, mine replacement becomes more important. Projects discovered in the 2020s should be operating or abandoned. The market will know whether automation, processing, and exploration offset grades and permitting. Central-bank behavior will have passed through multiple policy cycles.

The central scenario assumes mine supply near 4,000 tonnes and recycling near 1,700 tonnes. If output is materially lower while official and private allocations remain stable, the upper half of the range becomes more plausible. If supply is higher and portfolio allocations fall, the lower half dominates.

#Transaction structure by 2040

A mature market may include continuous fractional trading, standardized bar identity, and faster cross-border settlement. That can lower friction but also make risk-off selling faster. Improved liquidity does not determine direction; it increases transmission speed.

Back to top ↑

#21. Gold price forecast for 2045

#Central range: $4,200–$7,600 real 2026 dollars

The 2045 range reflects growing uncertainty about monetary institutions and mining. A quarter-century contains enough time for reserve currencies, fiscal norms, payment systems, and geopolitical alliances to change materially.

A price near $4,200 would represent long-run purchasing-power preservation with limited additional premium. A price near $7,600 would represent a larger allocation to gold but not a disorderly monetary system.

The physical model assumes recycling reaches 1,800 tonnes, mine output falls to 3,850 tonnes, and central-bank demand remains 600 tonnes. Those are not independent. A high price can raise recycling, improve mine economics, suppress jewellery, and reduce official buying simultaneously.

Back to top ↑

#22. Gold price forecast for 2050 and 2051

#Central 2051 range: $4,600–$8,800 real 2026 dollars

The 2050 and 2051 forecasts are best understood as regimes. By then, the exact number of dollars will depend heavily on cumulative inflation. At 2.5% annual inflation, the central 2051 real range is roughly $8,528–$16,312 nominal. The stress range converts to roughly $14,460–$29,660 nominal.

The central midpoint of $6,700 real represents about 1.6% annualized real appreciation from the model’s 2026 midpoint. That is materially below the historical nominal return since 1971 and intentionally conservative about extrapolating the post-Bretton Woods period.

#What “gold at $10,000” could mean

A $10,000 nominal price before 2051 may occur without the stress scenario if inflation is cumulative. A $10,000 real 2026-dollar annual average would be a much stronger statement, implying a large increase in purchasing power and likely a larger strategic allocation.

Search headlines often omit this distinction. Any serious gold price prediction for 2050 should specify currency, unit, price basis, averaging period, and inflation convention.

#What would make the 2051 central case wrong

The central case would be too high if trusted real returns remain strong, central banks reach target allocations early, gold loses relevance to other neutral reserve assets, mine output expands, and private allocation declines. It would be too low if sovereign debt stress, sanctions, reserve fragmentation, or a technological/custody shift produces a durable increase in desired gold holdings.

Back to top ↑

#23. Real versus nominal gold prices

#Inflation can create impressive nominal targets without real gains

At 2.5% inflation, the general price level rises about 85% over 25 years. A gold price that rises from $4,550 to $8,430 would roughly preserve purchasing power before storage, spreads, tax, and fees. This is why nominal targets must not be presented as automatic wealth creation.

#Three illustrative inflation conversions for the 2051 central midpoint

These are mathematical conversions, not forecasts of consumer prices. Actual inflation can vary across countries and products, and exchange rates can dominate local outcomes.

#Gold’s own holding costs matter

Physical gold can incur dealer spreads, storage, insurance, transport, assay, and tax. ETFs charge fees and have wrapper risks. Futures require margin and rolling. Jewellery includes workmanship and can be illiquid. A gross benchmark return is not a net investor return.

Back to top ↑

#24. Monitoring dashboard

A long-term forecast should change only when evidence changes. The following dashboard provides the most useful signposts.

#Monetary and financial variables

• 10-year and 5-year real yields in major currencies;
• expected and realized inflation;
• trade-weighted dollar and local gold prices;
• sovereign term premium and debt-service burden;
• central-bank balance sheets and liquidity conditions;
• banking stress, credit spreads, and market volatility.

#Official-sector variables

• annual central-bank net purchases in tonnes;
• number and geographic breadth of buyers;
• changes in target reserve allocations;
• domestic versus foreign custody;
• gold lending, swaps, and mobilization;
• coordinated or large official sales.

#Private-investment variables

• global ETF holdings and flows;
• OTC and exchange volume;
• COMEX positioning and leverage;
• bars-and-coins demand;
• retail premiums and dealer inventories;
• gold as a share of global financial assets.

#Physical variables

• mine production, grades, costs, and project pipeline;
• recycled supply and local selling behavior;
• jewellery tonnage and value;
• technology demand and thrifting;
• refiner utilization and treatment charges;
• bar availability across locations.

#Market-structure variables

• LBMA benchmark and Good Delivery changes;
• allocated/unallocated spreads and lease rates;
• vault holdings and transfer data;
• tokenized and fractional product regulation;
• provenance database adoption;
• sanctions and responsible-sourcing rules.

#Practical update triggers

A model review should occur before the scheduled date if any of the following persists for at least two quarters:

• central-bank demand moves below 300 tonnes annualized or above 1,200 tonnes;
• global ETF holdings change by more than 20% from the model date;
• 10-year U.S. real yields move below -1% or above 4%;
• annual mine guidance changes by more than 300 tonnes;
• recycled supply exceeds 2,000 tonnes before 2030;
• the benchmark trades outside the current scenario envelope for six months;
• major legal changes alter gold ownership, taxation, sanctions, or custody.

Back to top ↑

#25. Buyer playbook

#2026–2030: separate thesis from entry price

A buyer near a record should assume that 20–35% drawdowns are possible even in the central or upside regime. Staging purchases can reduce timing concentration, though it cannot guarantee a better result. Choose the instrument based on purpose:

• possession and low counterparty exposure: physical allocated metal;
• portfolio liquidity: ETF or vault account;
• tactical exposure or hedging: futures and options;
• cultural and wearable use: jewellery;
• collectible interest: numismatic market, evaluated separately from melt value.

Compare total acquisition cost, not just the quoted premium. Verify dealer reputation, title, buyback terms, authenticity, tax, storage, insurance, and jurisdiction.

#2031–2040: review the role, not only the price

A holding accumulated for crisis insurance may no longer fit if fiscal and geopolitical conditions improve. Conversely, a low allocation may need review if reserve fragmentation deepens. Rebalance against a written objective rather than momentum.

For institutional buyers, custody concentration becomes a major issue. Diversifying vault location can reduce one risk while increasing legal, operational, and transport complexity.

#2041–2051: distrust convenience without legal clarity

Fractional, tokenized, and continuously traded products will likely be common. Buyers should demand proof of assets, bar lists where applicable, redemption rights, audited reconciliation, legal segregation, and insolvency treatment. A perfect app does not cure weak title.

#Questions every buyer should answer

1. What problem is gold solving?
2. What is the intended holding period?
3. Is the exposure allocated, unallocated, derivative, or merely contractual?
4. What are all-in purchase and sale costs?
5. Who is the counterparty and custodian?
6. Where is the gold legally located?
7. How is authenticity verified?
8. What tax and reporting rules apply?
9. Can the position be sold during stress?
10. What evidence would cause the buyer to reduce the holding?

Back to top ↑

#26. Seller playbook

#Identify what is being sold

Before requesting quotes, determine:

• gross weight and unit;
• stated and likely purity;
• stones, solder, fillings, or non-gold components;
• brand, maker, age, rarity, and condition;
• bar or coin serials and packaging;
• provenance and purchase records;
• whether the item may exceed melt value.

#Compare the same settlement basis

One buyer may quote a percentage of spot before fees; another may quote a net amount after assay; a third may offer consignment. Put every offer into a common format:

Net proceeds = payable fine gold × reference price − all fees, tax, shipping, and delay cost

Record the benchmark time, currency, purity assumption, payable percentage, payment date, and recourse if the assay differs.

#Do not confuse spot, melt, and resale value

• Spot or benchmark: reference wholesale price.
• Gross melt value: benchmark value of estimated fine gold.
• Payable value: gross value multiplied by the buyer’s payable percentage.
• Net settlement: payable value after fees and deductions.
• Resale or collectible value: market value of the item as an item, which can be above or below melt.

#Choose timing according to need, not certainty

No seller knows the exact top. A holder who needs liquidity should compare the cost of waiting with the possibility of a higher price. A large seller can divide a sale, hedge, or negotiate refinery terms. A small seller should prioritize a transparent and reputable counterparty over a headline quote that cannot be verified.

#Preserve evidence

Photograph items, serial numbers, seals, scale readings, and packaging before transfer. Use insured shipping and tamper-evident packaging. Retain assay reports, receipts, and communications. For large lots, negotiate sampling and umpire-assay procedures before the buyer takes custody.

Back to top ↑

#27. Risks that would invalidate the model

#A new neutral reserve asset displaces gold

A widely trusted, politically neutral, highly liquid reserve instrument with credible scarcity and no sovereign issuer could reduce gold’s strategic role. It would need to survive crises and gain official acceptance, not merely attract speculative trading.

#Major monetary reform changes the unit of account

A formal gold link, commodity basket, central-bank digital settlement regime, or redesigned international reserve system could change demand and price formation in ways not captured here. A return to an official gold price would make free-market extrapolation especially unreliable.

#Governments restrict ownership or trade

Confiscation is not the only risk. Reporting, capital controls, windfall taxes, transaction taxes, import restrictions, export controls, cash limits, or sanctions can alter local prices and liquidity. Gold can rise globally while becoming difficult to monetize in one jurisdiction.

#Mine technology changes supply economics radically

A step change in exploration, recovery, seabed access, or processing could increase supply. Conversely, stronger environmental constraints or energy scarcity could lower it sharply.

#Recycling technology unlocks inaccessible stocks

Highly efficient, low-cost recovery from e-waste, tailings, or low-grade materials could add supply beyond the model. The effect would depend on collection systems and whether the recovered metal is sold or retained in closed loops.

#Central banks become sustained net sellers

A multi-year official-sector shift from buying to selling would directly challenge the central and stress scenarios. The reason would matter: restored confidence is bearish; emergency sales during crisis can be ambiguous.

#Gold loses its social and cultural demand base

A generational shift away from gold jewellery and physical savings could weaken the demand floor. The model assumes adaptation rather than disappearance. Evidence of broad and persistent abandonment would require lower long-term ranges.

#The current price contains a much larger bubble than assumed

The 2025–2026 rise may have pulled forward years of demand. If real yields remain high and momentum reverses, price can fall below the downside band temporarily. The model’s annual-average bands should not be treated as short-term support levels.

Back to top ↑

#28. Frequently asked questions

#Will gold reach $10,000 per ounce?

A $10,000 nominal price is plausible within the central scenario by 2051 under ordinary cumulative inflation. A $10,000 price in real 2026 dollars requires an outcome closer to the upper central or monetary-stress regime. The date, inflation basis, and averaging period matter.

#What is the gold price forecast for 2030?

The central scenario is $4,000–$5,900 per ounce in real 2026 dollars. The broader scenario envelope is $2,900–$7,800. These are annual-average regimes.

#What is the gold price forecast for 2040?

The central scenario is $4,000–$6,900 real. The full envelope is $2,300–$11,500.

#What is the gold price forecast for 2050 or 2051?

The 2051 central range is $4,600–$8,800 real 2026 dollars. The full envelope is $2,200–$16,000. At 2.5% inflation, the central nominal range is approximately $8,528–$16,312.

#Is gold guaranteed to rise with inflation?

No. Gold can underperform inflation for long periods. Inflation, real rates, currency, valuation, and investor positioning interact. Gold has historically preserved purchasing power over some long intervals, but entry price matters.

#Do high interest rates always make gold fall?

No. They raise opportunity cost, but gold can rise if buyers expect rates to fall, distrust the currency or sovereign issuer, seek crisis insurance, or if official-sector demand is strong.

#Will central banks keep buying gold?

The 2026 evidence is supportive, but not guaranteed. Survey intentions, actual purchases, and valuation effects must be distinguished. The central model assumes buying remains elevated but gradually declines.

#Can mine supply satisfy higher demand?

It can respond, but slowly. High prices improve economics while costs, grades, permitting, finance, infrastructure, and politics constrain new output. Existing above-ground gold and recycling are more immediate sources.

#Will recycled gold cap the price?

Recycling is an important buffer, not a fixed ceiling. Sellers may hold back if they expect higher prices, and some gold is culturally or strategically sticky. A very high price can nevertheless mobilize large volumes.

#Is jewellery a good way to invest in gold?

Jewellery can store value, but workmanship, retail margin, tax, fashion, condition, and resale channel affect returns. Its recovery value can be far below purchase price even when the gold benchmark rises.

#Are gold ETFs the same as owning physical gold?

No. They provide market exposure through a security with custody and fund structures. They can be highly liquid but do not usually give small holders direct possession or individual bar title.

#Is allocated gold safer than unallocated gold?

Allocated gold reduces some credit exposure by identifying specific bars. It can cost more and be less operationally flexible. Safety depends on title, custody, jurisdiction, insurance, audit, and legal terms.

#Are tokenized gold products safe?

Only if the underlying legal, custody, audit, redemption, and insolvency arrangements are sound. Blockchain records cannot compensate for missing or encumbered metal.

#What is the best way to sell gold?

Identify the item, test purity, calculate gross contained value, obtain multiple net quotes, and choose the market suited to bullion, jewellery, collectible, or industrial scrap. Use a documented and reputable counterparty.

#Why is a dealer’s offer below spot?

Spot is a wholesale reference. The dealer must account for authentication, assay, hedging, refining, inventory, payment, fraud, overhead, and profit. The fair question is whether the net spread is transparent and competitive.

#Could gold fall below $2,000 again?

A temporary nominal move cannot be ruled out in a severe liquidation or monetary-credibility boom, but it is below the long-term annual-average downside bands in this model. Such a decline would require a large compression in the current risk and allocation premium.

#Is gold a better investment than stocks or bonds?

That cannot be answered universally. Gold has no cash flow, while stocks and bonds can produce income. Gold may diversify issuer, currency, and crisis risk. The appropriate role depends on objectives, valuation, horizon, and constraints.

#How often will this forecast be updated?

At least annually, with an earlier review after material changes in real yields, central-bank behavior, mine supply, recycling, benchmark rules, or market regulation. Prior versions should remain public.

Back to top ↑

#29. Methodology, source hierarchy, and limitations

#Source hierarchy

The research prioritizes:

1. benchmark administrators and market-standard bodies;
2. central banks, the IMF, World Bank, U.S. Geological Survey, OECD, and United Nations bodies;
3. World Gold Council supply, demand, holdings, and market-structure research;
4. exchange contract specifications and clearing data;
5. peer-reviewed academic literature;
6. company and industry material for narrowly attributed operational facts.

The World Gold Council is an industry membership organization. Its data are widely used and often produced with Metals Focus, but its institutional perspective and disclaimers should be understood. The paper cross-checks macro and reserve claims with the IMF, World Bank, FRED, LBMA, CME, OECD, USGS, and academic research where possible.

#Model units

• Supply and absorption: metric tonnes.
• Prices: annual-average U.S. dollars per troy ounce.
• Primary price basis: inflation-adjusted 2026 dollars.
• Nominal conversions: illustrative constant inflation assumptions.
• Above-ground stock: prior stock plus modeled mine production; recycling does not create new stock.

#The institutional residual

The model field institutional_etf_otc_absorption_required_t equals:

total supply − jewellery − technology − central-bank demand − bar-and-coin demand

It is a balancing category. It can include ETFs, OTC investment, exchange and dealer inventories, institutional vaulted holdings, and statistical differences. It should not be interpreted as a direct prediction of disclosed ETF flows.

#Price construction

Price bands are judgmental regimes informed by:

• the 2026 starting valuation;
• physical model paths;
• above-ground stock;
• desired allocation relative to financial wealth;
• real yields and currencies;
• official-sector demand;
• macro and geopolitical risk;
• mine and recycling response;
• historic drawdowns and structural breaks.

The ranges are not generated by ordinary least squares, machine learning, Monte Carlo simulation, or a claim of statistical confidence. A purely statistical model trained on the post-1971 period would inherit regime assumptions that may not survive to 2051.

#Limitations

The model does not predict:

• daily, monthly, or exact year-end prices;
• exchange rates or country-specific local gold prices;
• inflation paths;
• tax, legal, or sanctions outcomes;
• mine-by-mine production;
• discoveries, grades, or technological breakthroughs;
• central-bank transactions by institution;
• ETF product flows by issuer;
• jewellery taste or cultural change by country;
• dealer spreads, premiums, or scrap payables;
• wars, confiscation, capital controls, or monetary reform.

The three scenarios are not exhaustive. Reality can combine restored monetary credibility in one region with severe stress in another. Gold is global, so the marginal price can reflect multiple regimes at once.

#Versioning rule

A substantive revision must:

• preserve model version 1.0;
• publish a dated change log;
• identify the changed evidence and assumptions;
• regenerate the published tables and figures;
• explain why price bands moved;
• compare prior ranges with subsequent annual averages;
• retain the original publication date and update dateModified.

A forecast gains credibility by exposing mistakes, not by deleting them.

Back to top ↑

#30. Conclusion

Gold’s next 25 years will not be determined by one mine, one inflation rate, one reserve currency, or one technology. The market is too large, too global, and too deeply embedded in official reserves, private wealth, jewellery, and financial infrastructure for a single-factor forecast.

The central case is managed fragmentation. It assumes that the international monetary system remains functional but less trusted and less unified than it was during the peak of post-Cold War globalization. Central banks continue to hold and buy gold, though at a slower pace than the extraordinary early 2020s. Private investors maintain a meaningful allocation. Mine output reaches a plateau and then becomes harder to replace. Recycling grows. Jewellery adapts. Digital access expands, while title, custody, jurisdiction, and provenance become more—not less—important.

Under that case, the annual-average gold price in real 2026 dollars can remain broadly near today’s purchasing power through the early 2030s and rise gradually to $4,600–$8,800 by 2051. That is not a promise of exceptional investment returns from the 2026 starting point. It is a forecast that gold preserves a durable monetary and portfolio role.

The downside case is credible if real monetary trust is restored. The upside case is credible if the world assigns more balance-sheet space to an asset without an issuer. Both outcomes are possible, and both contain severe interim volatility.

For buyers, the essential decision is not only price; it is wrapper, title, custody, liquidity, and purpose. For sellers, the essential decision is not only spot; it is identification, purity, market channel, payable percentage, evidence, and net settlement. Technology will improve the transaction layer, but it will not remove those fundamentals.

The forecast should therefore be used as a map of conditions. Watch the evidence. Preserve the old version. Update the assumptions. Refuse the false precision of a single 2051 number.

---

Back to top ↑

#References

Baur, Dirk G., and Brian M. Lucey. 2010. “Is Gold a Hedge or a Safe Haven? An Analysis of Stocks, Bonds and Gold.” Financial Review 45 (2): 217–229. https://doi.org/10.1111/j.1540-6288.2010.00244.x

Baur, Dirk G., and Thomas K. McDermott. 2010. “Is Gold a Safe Haven? International Evidence.” Journal of Banking & Finance 34 (8): 1886–1898. https://doi.org/10.1016/j.jbankfin.2009.12.008

CME Group. 2026a. “Gold Futures Contract Specs.” https://www.cmegroup.com/markets/metals/precious/gold.contractSpecs.html

CME Group. 2026b. “Micro Gold Overview.” https://www.cmegroup.com/markets/metals/precious/e-micro-gold.html

CME Group. 2026c. “FAQ: 1-Ounce Gold Futures.” https://www.cmegroup.com/articles/faqs/faq-1-oz-gold-futures.html

CME Group. 2026d. “Gold Futures Overview.” https://www.cmegroup.com/markets/metals/precious/gold.html

Erb, Claude B., and Campbell R. Harvey. 2013. “The Golden Dilemma.” NBER Working Paper 18706. https://www.nber.org/papers/w18706

Federal Reserve Bank of St. Louis. 2026a. “10-Year Treasury Inflation-Indexed Security, Constant Maturity.” FRED series DFII10. https://fred.stlouisfed.org/series/DFII10

Federal Reserve Bank of St. Louis. 2026b. “10-Year Treasury Constant Maturity Rate.” FRED series DGS10. https://fred.stlouisfed.org/series/DGS10

International Monetary Fund. 2026a. “IMF Data Brief: Currency Composition of Official Foreign Exchange Reserves, First Quarter 2026.” July 1. https://data.imf.org/en/news/imf%20data%20brief%20july%201

International Monetary Fund. 2026b. “COFER Frequently Asked Questions.” https://data.imf.org/en/Datasets/COFER/Frequently-Asked-Questions

International Monetary Fund. 2026c. “World Economic Outlook Update, July 2026: Global Economy in Crosscurrents of War and Technology.” https://www.imf.org/en/publications/weo/issues/2026/07/08/world-economic-outlook-update-july-2026

LBMA. 2025a. “Laying the Digital Foundation: The GBI Database and Beyond.” Alchemist 115. https://www.lbma.org.uk/alchemist/issue-115/the-gbi-database-and-beyond

LBMA. 2026a. “LBMA Precious Metal Prices.” https://www.lbma.org.uk/prices-and-data/lbma-precious-metal-prices

LBMA. 2026b. “LBMA Daily Trade Reporting Data.” https://www.lbma.org.uk/prices-and-data/lbma-daily-trade-reporting-data

LBMA. 2026c. “Clearing Data.” https://www.lbma.org.uk/prices-and-data/clearing-data

LBMA. 2026d. “About Loco London.” https://www.lbma.org.uk/market-standards/about-loco-london

LBMA. 2026e. “Responsible Sourcing.” https://www.lbma.org.uk/responsible-sourcing

LBMA. 2026f. “Gold Price: LBMA Snapshot Survey of Professional Analysts.” https://www.lbma.org.uk/articles/gold-price-lbma-snapshot-survey-of-professional-analysts

LBMA and World Gold Council. 2022. “LBMA and WGC Launch Gold Bar Integrity Programme.” https://www.lbma.org.uk/articles/lbma-and-wgc-launch-gold-bar-integrity-programme

Minamata Convention on Mercury. 2026. “Artisanal and Small-Scale Gold Mining.” https://minamataconvention.org/en/topics/artisanal-and-small-scale-gold-mining/insights

O’Connor, Fergal A., Brian M. Lucey, Jonathan A. Batten, and Dirk G. Baur. 2015. “The Financial Economics of Gold—A Survey.” International Review of Financial Analysis 41: 186–205. https://doi.org/10.1016/j.irfa.2015.07.005

OECD. 2016. OECD Due Diligence Guidance for Responsible Supply Chains of Minerals from Conflict-Affected and High-Risk Areas. Third edition. https://www.oecd.org/en/publications/oecd-due-diligence-guidance-for-responsible-supply-chains-of-minerals-from-conflict-affected-and-high-risk-areas_9789264252479-en.html

UNITAR and ITU. 2024. “Global E-Waste Monitor 2024: Electronic Waste Rising Five Times Faster than Documented E-Waste Recycling.” https://unitar.org/about/news-stories/press/global-e-waste-monitor-2024-electronic-waste-rising-five-times-faster-documented-e-waste-recycling

U.S. Geological Survey. 2026. Mineral Commodity Summaries 2026. https://doi.org/10.3133/mcs2026

World Bank. 2026. “Precious Metals Retreat from Record Highs.” https://blogs.worldbank.org/en/opendata/precious-metals-retreat-from-record-highs

World Gold Council. 2024. “Gold’s Long-Term Expected Return (GLTER).” https://www.gold.org/goldhub/research/golds-long-term-expected-return-glter

World Gold Council. 2026a. “How Much Gold Has Been Mined?” https://www.gold.org/goldhub/data/how-much-gold

World Gold Council. 2026b. “Gold Demand Trends: Q4 and Full Year 2025.” https://www.gold.org/goldhub/research/gold-demand-trends/gold-demand-trends-full-year-2025

World Gold Council. 2026c. “Gold Mid-Year Outlook 2026: Point Break.” https://www.gold.org/goldhub/research/gold-mid-year-outlook-2026

World Gold Council. 2026d. “Central Banks Set to Step Up Gold Buying over the Next Year.” June 16. https://www.gold.org/news-and-events/press-releases/central-banks-set-step-gold-buying-over-next-year

World Gold Council. 2026e. “Gold Demand Trends Q2 2026: Central Banks.” https://www.gold.org/goldhub/research/gold-demand-trends/gold-demand-trends-q2-2026/central-banks

World Gold Council. 2026f. “Gold ETF Flows: July 2026.” August 6. https://www.gold.org/goldhub/research/gold-etfs-holdings-and-flows/2026/08

World Gold Council. 2026g. “Gold Demand Trends Q2 2026: Supply.” https://www.gold.org/goldhub/research/gold-demand-trends/gold-demand-trends-q2-2026/supply

World Gold Council. 2026h. “Gold as a Strategic Asset: 2026 Edition—Return.” https://www.gold.org/goldhub/research/relevance-of-gold-as-a-strategic-asset/key-attributes-return

World Gold Council. 2026i. “Gold Demand Trends: Q2 2026.” https://www.gold.org/goldhub/research/gold-demand-trends/gold-demand-trends-q2-2026

World Gold Council. 2026j. “Gold Demand Trends Q4 and Full Year 2025: Supply.” https://www.gold.org/goldhub/research/gold-demand-trends/gold-demand-trends-full-year-2025/supply

World Gold Council. 2026k. “Gold Market Primer: Market Size and Structure.” https://www.gold.org/goldhub/research/market-primer/gold-market-primer-market-size-and-structure

---

#About this publication

All Metal Prices produces long-form reference material about metal prices, markets, valuation, buying, selling, recycling, and historical development. The analysis distinguishes reported facts from All Metal Prices model outputs. It does not claim undisclosed credentials, proprietary market access, or certainty about future prices.

Corrections and updates: Material factual errors should be corrected promptly with a dated note. Substantive forecast changes require a new model version and preserved prior files.

Commercial disclosure: This article may appear on a commercial informational website. Any future advertising, affiliate relationships, lead-generation arrangements, or sponsored content should be disclosed separately and must not determine the forecast ranges or source selection.