Gold touched an all-time intraday high of $5,589 per ounce on January 28, 2026. Since then it has corrected roughly 25 percent, trading between $4,121 and $4,143 in the week of July 6-10, 2026, per Trading Economics. That correction has split Wall Street: some banks still see $6,000 gold, others just cut targets by double digits. This article collects every major bank forecast for 2026 and 2027, dated, so you can see not just where the targets sit but how fast they move.
One thing before the numbers. Every forecast below belongs to the named bank as of the date shown. None is our prediction, none is investment advice, and banks revise these targets constantly, sometimes within weeks.
The setup matters more than any single target. Gold returned about 27 percent in 2024 (Statista) and roughly 65 percent in 2025, its best calendar year since 1979 (Visual Capitalist). The rally peaked at $5,589 intraday on January 28, 2026. Then came the correction: about 25 percent from the high, landing in the low $4,100s by early July. Even after that drawdown, gold is still up around 23 percent year over year.
A 25 percent correction after a 65 percent year is not unusual historically, but it changes the forecasting math. Targets written in January assumed momentum; targets written in June and July have to explain a drawdown. That is why the numbers below cluster into two groups: those published before the correction deepened, and those cut after it.
All figures are per troy ounce, attributed to the bank and the month published or reported. Where a bank has revised, we show the revision, because the revision history tells you more than the target.
JPMorgan carried the boldest headline target of the cycle: $6,000 by end-2026, with research earlier in the year pointing toward roughly $6,300 by end-2027. The bank reaffirmed the $6,000 call as recently as May and June 2026. Then, per JPMorgan, July 2026, it slashed the year-end 2026 target to $4,500, citing weaker-than-expected demand from key sectors, while keeping a constructive view into 2027 on continued central bank purchases. From $6,000 to $4,500 in a matter of weeks: remember that the next time a single target feels like certainty.
Per Goldman Sachs, January 2026, analysts Daan Struyven and Lina Thomas raised the bank's end-2026 target to $5,400 from $4,900. Per Goldman Sachs, June 2026, the same team cut it back to $4,900, pointing to fading ETF inflows, including the first monthly Asian ETF outflow since August 2025, and the removal of expected 2026 rate cuts from the bank's Fed forecast. Goldman has also flagged a downside scenario around $4,400 by year-end if the Fed were to raise rates.
Per Bank of America, July 2026, the bank cut its 2026 average gold price forecast by 14 percent to $4,360, citing a more hawkish Federal Reserve. At the same time, BofA has said its $6,000 twelve-month target remains in reach once the tightening cycle ends, and commodity strategist Michael Widmer has flagged an extreme-demand scenario in which gold reaches $8,000 by 2027: a lower near-term average, a high conditional ceiling.
Per Citi, June 2026, the bank cut its zero-to-three-month gold target to $4,000 from $4,300, warning of limited near-term upside as macro conditions improve. Citi's published scenario work has included a bull case, assigned roughly 30 percent probability, of $5,000 by end-2026 and $6,000 by end-2027, against a bear range of $3,600 to $3,800. Citi is the clearest example of a bank publishing a probability tree rather than a single number.
Per UBS, reported in mid-2026, the bank raised its twelve-month target, looking out to mid-2027, to $5,200, roughly 30 percent above July 2026 spot, with an upside scenario of $7,200 if geopolitical risks escalate materially. UBS expects annual central bank buying to hold in the 750 to 1,000 tonne range for 2026.
Per HSBC, July 2026, the bank lowered its 2026 and 2027 forecasts, citing a hawkish shift in US monetary policy expectations and a stronger dollar. Its 2027 average forecast moved to $4,925 from $5,000, with a 2027 year-end figure around $5,025. HSBC still argues that structural fiscal and geopolitical risks support a resumption of the rally over time.
Strip out the price numbers and the banks are mostly arguing about the same four forces.
Central bank buying. Central banks bought 863 tonnes of gold in 2025, led by Poland with 102 tonnes, per the World Gold Council. The WGC's 2026 survey found a record 45 percent of central banks plan to increase gold reserves, and 89 percent expect global central bank holdings to rise. The WGC forecasts roughly 850 tonnes of official-sector buying in 2026. This is the floor under most bull cases.
ETF flows. Gold ETFs took in a record $89 billion, or 801 tonnes, in 2025, and January 2026 set a single-month record at $18.7 billion of inflows, per the WGC. But March 2026 saw record Western ETF outflows, only partly offset by Asian inflows. Flows drove the rally up, and flows are what the June-July target cuts are really about.
Rates and the dollar. The 10-year Treasury yielded 4.56 percent as of July 10, 2026. Goldman, BofA, and HSBC all tied their recent cuts to a more hawkish Fed under new Chair Kevin Warsh and a firmer dollar, since higher real yields raise the opportunity cost of a zero-yield asset. This is the most cited variable in the revision notes.
Geopolitics. The Iran conflict that began February 28, 2026 reminded markets why gold carries a crisis premium; the IEA called the Strait of Hormuz disruption the largest oil-supply disruption in history. UBS's $7,200 upside scenario is explicitly a geopolitical-escalation case. For a fuller breakdown, see our guide to what actually drives gold prices.
A credible roundup has to include the argument against. First, the correction is not hypothetical: it already happened, roughly 25 percent, and anyone who bought the January high is sitting on a meaningful drawdown. Second, Western investors have shown they will leave: the March 2026 record Western ETF outflows prove the flow picture can reverse quickly. Third, the rate backdrop turned. If the Fed stays hawkish and real yields keep rising, the opportunity-cost argument works against gold, which is why Goldman published a $4,400 downside case and Citi's bear range sits at $3,600 to $3,800.
What would break the bull thesis entirely? Central bank buying slowing well below the 750 to 1,000 tonne range, sustained positive real rates with a strong dollar, and de-escalation of the geopolitical risks currently priced in. The honest summary: the structural story remains intact per the banks that follow it, while the cyclical, flow-driven story has weakened since January.
Look back at the revision history above. JPMorgan: $6,000, reaffirmed, then $4,500, all inside about six weeks. Goldman: up to $5,400 in January, back to $4,900 in June. Citi: $4,300 to $4,000 in one note. These are serious research teams; the point is not that they are bad at forecasting. The point is that a price target is a snapshot of a model's inputs on the day it was published, and the inputs change weekly.
The forecast is not the edge. What you do when the forecast is wrong, and it will be wrong, is the edge.
Professional allocators generally do not trade bank targets. They treat them as a map of consensus and disagreement, then run a process that does not depend on any single number being right. That is the core argument for systematic, rules-based gold strategies: entries, exits, and risk are defined in advance, so a $1,000 revision in some bank's target never forces an emotional decision.
Three tools let investors use this information without betting on any single target.
Allocation bands. Mainstream guidance for gold allocations typically falls between 5 and 15 percent of a portfolio, with 10 percent the most cited figure; World Gold Council research found allocations of 4 to 15 percent improved risk-adjusted returns historically. Pick a band that fits your situation and stay inside it regardless of headlines. Our guide on how much gold belongs in a portfolio covers the research in depth.
Rebalancing. A band plus rebalancing converts volatility into discipline. After a +65 percent year, rebalancing forced trims near the highs; after a 25 percent correction, it flags whether you are under your band. No prediction required, just arithmetic. The same logic applies to the broader gold versus stocks decision.
Rules-based exposure. For investors who want tactical exposure rather than a static allocation, the alternative to discretionary forecasting is a systematic approach: predefined signals, exits, and position sizes. The value is not a better crystal ball. It is that decision quality stays constant whether gold is at $5,589 or $4,100.
To restate the compliance point plainly: every forecast here is the published view of the named bank as of the date shown, not ours. Bank forecasts are frequently revised, sometimes drastically, and nothing in this article is investment advice or a recommendation to buy or sell gold.
Nobody knows, and the banks disagree with each other and with their own prior notes. As of July 2026, published end-2026 views include JPMorgan at $4,500 (cut from $6,000 in July 2026), Goldman Sachs at $4,900 (June 2026), and BofA with a $4,360 average forecast for the year (July 2026), while BofA and UBS keep higher 12-month targets of $6,000 and $5,200. Treat all of these as dated snapshots, not promises.
Published 2027 views include JPMorgan research earlier in 2026 pointing toward roughly $6,300 by end-2027, Citi's bull case of $6,000 by end-2027 (assigned about 30 percent probability), UBS at $5,200 for mid-2027, HSBC around $5,025 for end-2027 (July 2026), and a BofA extreme scenario of $8,000. The spread itself is the message: 2027 is deeply uncertain even to the largest research desks.
Over 2024-2025 the main drivers were record central bank buying (863 tonnes in 2025 per the World Gold Council), record ETF inflows ($89 billion in 2025), expectations of easier monetary policy, and geopolitical risk, including the 2026 Iran conflict. Note that in mid-2026 gold has actually been correcting, as some of those drivers, especially Western ETF flows and rate-cut expectations, have weakened.
It can fall further, and it has already corrected roughly 25 percent from the January 2026 high of $5,589. Published bear scenarios include Goldman's $4,400 case if the Fed hikes and Citi's $3,600 to $3,800 range. A full crash scenario would likely require central bank buying to stall alongside sustained high real yields. No one can rule that in or out, which is why position sizing matters more than conviction.
That is an allocation question, not a timing question, and it depends on your situation. Research cited by the World Gold Council found 4 to 15 percent allocations historically improved risk-adjusted returns, and a rebalancing discipline handles the entry-timing problem mechanically. Investors who feel they must time the entry are implicitly forecasting, which is exactly what this article suggests treating with caution. This is education, not advice; consult a qualified professional about your own portfolio.