Gold Price Forecast: The Road to $10,000 (2026 Outlook)
On 19 May 2026, ex-Goldman commodities strategist Jeffrey Currie set out a two-part forecast for gold: a near-term pullback toward $4,000 an ounce, then a structural climb to $10,000. From a spot price of around $4,500, that long-run target implies roughly +122% of upside. The figure that gets the headlines is $10,000. The figure that matters for an investor is the gap between today’s price and that target, and the supply-side data that underwrites it.
GBP equivalents shown at 1 USD ≈ £0.74.
The forecast in one paragraph
Currie, formerly head of commodities research at Goldman Sachs for fifteen years and now at Carlyle Group, has been broadly bullish on gold for two decades. He opened a tactical short in March 2026 from higher levels and remains short into mid-May, expecting another leg lower toward $4,000 an ounce (roughly £2,960) before sentiment resets. He has stated he will return long when central banks turn dovish, which he expects once the energy crunch starts to bite growth. The long-run target he is working toward is $10,000 an ounce, equivalent to roughly £7,400 at current exchange rates.
Source: Mining.com, 19 May 2026. GBP equivalents at 1 USD ≈ £0.74.
Cycle vs structure
Why he expects a pullback first
The near-term pressure on the gold price is sentiment and flow, not fundamentals. Sovereign buying has been the single largest pillar of the gold bid since 2022. When the marginal sovereign turns net seller, the bid thins out before structural demand catches up again.
Turkey is the named example. The Turkish central bank, until recently a structural accumulator, has become a forced seller to fund higher energy import bills. Roughly 120 tonnes of Turkish reserves have been off-loaded in the relevant window. The wider read-through is that energy-exposed economies with thin reserves become forced sellers in this part of the cycle. If it is a one-off, the thesis breaks. If it spreads to other thinly-reserved central banks, it sticks.
Spot gold has slipped to around $4,500, year-to-date gains are around +5% versus a near-50% surge earlier in the cycle. Currie expects another leg lower toward $4,000 before sentiment resets. He has been clear that the position is a tactical bet, not a thesis change.
Why he still sees $10,000
The long-run case rests on two pillars: a starved supply side, and a brand-new demand source layered on top.
Supply side: a decade of underinvestment
Refinery investment sits at a ten-year low. Upstream oil and gas capex is 35% below its 2015 peak. The top 20 miners are spending 40% less than at the 2012 cycle high. Aged supply takes years to build back. The price floor under physical gold is, in the long run, the cost of replacing it, and that floor has been rising for a decade.
| Supply indicator | Level |
|---|---|
| Refinery investment | 10-year low |
| Upstream oil & gas capex vs 2015 peak | −35% |
| Top 20 miners’ capex vs 2012 cycle high | −40% |
Demand side: AI infrastructure as the new structural buyer
The Magnificent 7 plus Oracle alone will spend roughly $820 billion (£607bn) on capital expenditure in 2026, more than Germany’s entire annual capital formation. Commodities sit underneath every gigawatt of new compute: copper for power and cooling, silver and palladium for connectors and catalysts, and gold for electronics and as the reserve asset central banks hold against the dollars being printed to finance all of it.
“The most asymmetric trade in modern financial history.”
Jeffrey Currie · 19 May 2026
Currie’s historical analogues are Vietnam, 9/11, and the Iraq War, points where supply was already thin and a new demand source layered on top.
Where central banks actually stand
Turkey is the headline seller, not the whole picture. Central banks bought a net 863 tonnes in 2025, the fourth straight year above the multi-decade norm. China extended its disclosed accumulation. The United States, the world’s largest holder at 8,133 tonnes, has not sold. The structural picture is intact even with Turkey’s exit.
Source: World Gold Council, Q1 2026 Central Bank Gold Survey; IMF International Financial Statistics reserve totals.
How the major banks frame it
Currie’s $10,000 target is the most aggressive published view, but the direction of travel across the major desks is the same. The current 12 to 24 month range across J.P. Morgan, Deutsche Bank, and Goldman Sachs is approximately $5,400 to $8,500 an ounce (roughly £4,000 to £6,290). Currie is more cautious in the near term and more aggressive on the long view. None of the published views are short the structural case.
For investors, that consensus matters more than any single forecast. The disagreement between desks is about path and timing. The agreement is about direction. Multi-year price targets above today’s spot are no longer outliers.
For UK investors
What this means for UK investors
A long-cycle forecast is only useful if the holding vehicle is built for the cycle. The mismatch UK investors run into is buying long-cycle exposure through short-cycle, tax-inefficient wrappers. Three considerations matter more than usual when the thesis is measured in years.
Legal tender Britannias and Sovereigns are CGT-exempt under TCGA 1992 s21(1)(b), confirmed by HMRC manual CG78305. Bars are not.
On £100,000 of bars that doubles, a higher-rate taxpayer owes roughly £18,000 of CGT. On coins: zero. See the CGT on gold guide.
ETFs are claims, not metal. Fine for tactical exposure inside an ISA or SIPP. Less suited to the part of a portfolio designed to behave like gold when systems are under stress.
A multi-year thesis held through a custodial structure that depends on the issuer remaining solvent is a different bet than holding the metal.
Graded coins (NGC PF70, PCGS PR70) carry numismatic premium harder for the market to commoditise. Mintage is fixed, and top-grade populations only shrink as coins enter long-term storage.
Read the graded coins guide for how grading affects long-term value.
Counter-view
Risks and what would break the thesis
An honest read of any forecast includes the conditions that would falsify it. For Currie’s $10,000 target, the main risks are:
Sustained central bank net selling. If Turkey’s forced selling spreads to other emerging market central banks (the obvious candidates being any large reserve holder with concentrated energy import exposure), the bid that has underpinned the price since 2022 weakens before structural demand catches up. The near-term floor moves down before the long-run target comes into view.
A genuine break in AI capex. If the $820bn capex cycle stalls (regulatory action, an AI demand disappointment that hits hyperscaler earnings, or a financing squeeze), one of the two pillars under the long-run case weakens. The supply pillar remains intact, but the upside compression is more modest.
A faster supply response than expected. Mining capex can re-accelerate, particularly if prices stay elevated for long enough to fund new projects. The historical lag is 7 to 10 years from greenfield discovery to first production, but brownfield expansions can come on faster.
None of these scenarios are short the structural case for gold. They are reasons the price might not reach $10,000 on the timeline Currie suggests. The path is uncertain. The direction is not particularly contested across published forecasts.
The real story behind the headline number
Two previous long-cycle rallies, and where Currie sees the third going
Annual averages, USD per ounce. Dashed line shows Currie’s expected pullback to ~$4,000 then his long-run $10,000 target. Sources: World Gold Council, LBMA, Mining.com. Past performance does not guarantee future returns; projections are not promises.
The figure that gets the attention is $10,000. The figure that actually matters for an investor is two-sided: the supply discipline that has been building for a decade, and the demand source that has been added on top in the last three years. Strip out the specific target, and the underlying picture is the same one the World Gold Council, J.P. Morgan, Deutsche Bank, and Goldman Sachs have been pricing in. Currie’s contribution is to be louder about it.
For a UK investor with a multi-year horizon, the practical questions are: what format, what wrapper, and at what allocation. UK legal tender coins answer the first two cleanly. The third is a personal call, but the historical record from gold’s previous long-cycle rallies (1971-1980, 2001-2011) is that the investors who benefitted were the ones who owned the asset before the consensus arrived, in a format that let them keep what the price gave them.
Press coverage
Client Elaine Pickin’s £120,000 graded gold portfolio, with a £16,000 unrealised gain inside months and entirely CGT-free, was featured in The Telegraph in December 2025. Coverage of the tax structure that makes UK legal tender coins distinctive has appeared across the national press.
Frequently Asked Questions
Will gold really reach $10,000 an ounce?
Jeffrey Currie’s $10,000 target is the most aggressive published forecast from a major-bank strategist as of mid-2026. J.P. Morgan, Deutsche Bank and Goldman Sachs currently model a 12 to 24 month range of roughly $5,400 to $8,500 (about £4,000 to £6,290). Whether the long-run target is reached depends on the trajectory of central bank buying, AI-driven commodity demand, and the supply response from miners. The direction across published forecasts is broadly aligned. The path and timing are contested.
Why does Currie expect a pullback first?
The near-term pressure is flow-driven, not fundamental. Turkey’s central bank has turned net seller to fund higher energy import costs, off-loading roughly 120 tonnes. That removes part of the structural bid that has supported the price since 2022. Currie expects the price to drift toward $4,000 before sentiment resets and central banks turn dovish enough for him to close the short and return long.
What is the UK gold price forecast in pounds?
Translated at the current rate of roughly 1 USD ≈ £0.74, the published targets in sterling are: near-term floor ≈ £2,960, current spot ≈ £3,330, 12-24 month range across J.P. Morgan, Deutsche Bank and Goldman Sachs ≈ £4,000 to £6,290, Currie’s long-run target ≈ £7,400. The sterling figures will drift with the exchange rate.
Is gold a good investment for UK investors right now?
“Right now” depends on the alternative. Cash held in a UK current account in April 2014 had lost roughly 25% of its purchasing power by April 2026. Gold, over the same period, gained roughly four-fold in sterling terms. The case for an allocation to gold for a UK investor sits on three legs: it is a long-cycle inflation hedge with an underlying structural bid, the UK legal tender format is CGT exempt, and the holding does not depend on a custodian or counterparty staying solvent. For more on how gold sits in a wealth-protection allocation, see our guide to gold as wealth protection.
Why gold coins and not gold bars if I expect the price to rise?
The metal exposure is similar. The tax treatment is not. UK legal tender coins (Royal Mint Britannias and Sovereigns) are exempt from Capital Gains Tax. Gold bars are not. On a thesis that involves the price doubling or more from today’s spot, the difference between paying 0% CGT and 18-24% CGT on the gain is the difference between keeping the upside and giving away roughly a fifth of it. See our Sovereign vs Britannia guide for the choice between the two main UK coin formats.
What could break the $10,000 thesis?
Three scenarios. First, sustained central bank net selling beyond Turkey, particularly if it spreads to other energy-exposed reserve holders. Second, a genuine break in the AI capex cycle that removes the new structural demand layer. Third, a faster supply response from miners than the historical 7-10 year lag from discovery to production. None of these scenarios are short the structural case for gold. They affect the timing and the peak, not the direction.
How much of a portfolio should be in gold given this forecast?
There is no single right answer. The World Gold Council’s portfolio research suggests 4-15% across a diversified portfolio depending on risk profile. Ruffer Investment Company has historically run gold allocations of 5-15% across its multi-asset strategies. UK investors prioritising CGT efficiency commonly allocate 20-30% into UK legal tender coins, because the exemption shields the entire gain from tax regardless of size. Higher allocations are typically wealth preservation focused: inheritance recipients, pension drawdown stages, or any investor where tax minimisation is the priority for the bulk of the portfolio.
This guide is for informational purposes only and does not constitute financial, tax or investment advice. Physical gold investment is not regulated by the FCA and is not covered by the Financial Services Compensation Scheme. Speak to a regulated adviser for personalised recommendations.
Get the Gold Investment Guide
Book a call with one of our friendly specialists to get any of your questions answered.
