Gold Price Prediction: Sourced Targets for 2026 and 2027

Reviewed by the Gold & Silver Timing Guide Editorial Team · Updated November 3, 2026 at 16:50 UTC

Quick answer: Nobody can tell you where gold goes next, and this page does not pretend to. What it offers is a short list of targets that named publishers have printed, each with the date of the page we read, set beside the live price so you can see how far apart they sit. Every row is secondary: a news site, a news flash or a dealer's blog reporting what a bank or a poll said, never the bank's own research page. If you arrived asking should I buy gold or silver, a published target is one input at most, and the sections below explain why it is a weak one.

Left out on purpose:

• J.P. Morgan gold: the figures on the pages read disagree with each other, so none is printed
• Reuters polls: the Reuters page itself could not be opened, so its numbers appear only through the dealer compilation rows

What the table above is, and what it is not

Each row names the publisher of the number, the page that reported it, the date printed on that page, the horizon the target covers and whether the figure is a bank's own view or a poll. We read every page on November 3, 2026, and that is the date to keep in mind when you see a figure. The rows are a record of what was published. They are not an average, because we did not average anything. They are not a consensus, because the rows were not chosen to be representative. They are not our forecast, because we do not make one. And they are not a ranking of who is most likely to be right, because we did not follow any publisher's record.

The table is also not complete. It holds the figures we could read and stand behind. Figures for one large bank differed between the pages we read, with a gap too wide to be explained by timing alone, so we print none of them rather than pick a side. A poll of analysts dated 27 October 2026 appears below only as a dealer's compilation lists it, because the page of the news agency that ran the poll could not be opened on our side. Pages published before 2026 were left out, and so was a dealer blog that printed no date at all. A table that omits things should say so, and this is the list.

The compilation's own line on that poll reads $4,495 for the 2026 average and $4,650 for 2027, as the dealer page, updated 2026-11-01, listed them. Treat those two numbers with the care you would give any figure relayed twice.

How to read a target

A gold target is a sentence with four parts: who said it, when, what period it covers, and what number it names. Most of the confusion around forecasts comes from skipping the middle two.

The horizon. Some rows are a single level for the end of a year. Some are an average over a quarter. One row gives two levels for two different windows, a nearer one for the next three months and a later one for six to twelve months out. Bank of America's row in the compilation is an average for the fourth quarter of 2026 of about $4,000, with a separate figure for the first quarter of 2027. An average over a quarter and a year-end level can both be correct while sitting far apart, because they ask different questions. If you line up the numbers without lining up the horizons, the table looks more contradictory than it is.

The issue date versus the page date. The date we print is the date on the page we read. The date the bank issued its view can be earlier, sometimes by months. The same dealer compilation, updated on 2026-11-01, prints a UBS figure of $4,600 for the end of 2026 with an issue date in July. A figure that old has lived through a lot of price movement. The Wells Fargo row carries a further wrinkle: the news flash is dated 2026-08-17, yet its text refers to September, so the page contradicts itself on the month, and we flag that rather than guess.

Bank versus poll. A bank's target is one team's view. A poll figure is the middle of many views, collected at one moment. The Reuters entry in the table is a poll, so it answers a different question from Goldman Sachs' own end-of-year figure.

Secondary versus primary. We read no bank research note and no news-agency original for this table. We read pages that quote them. A quoting page can shorten, round or mislabel. That is the reason every row says secondary and the reason you should treat a figure as a pointer to go and check, not a fact to build on.

Where the 2026 rows sit relative to each other

Reading the rows for the end of 2026 side by side, without averaging them, shows a spread. At the low end is Bank of America's fourth-quarter average of about $4,000. Morgan Stanley's fourth-quarter figure is $4,450, UBS has $4,600 for year-end, and Goldman Sachs and Standard Chartered each list $4,650, though Goldman's is an end-of-year fair value and Standard Chartered's is a fourth-quarter average. Citi lists $4,800 for the nearest three months. At the top, Wells Fargo Investment Institute gives a range of $4,900 to $5,100 for year-end 2026.

Notice what that spread is. It is the distance between forecasters who are all looking at the same market on roughly the same weeks. It is wide because the future is not known, not because someone is careless. For 2027 the rows are fewer. Goldman Sachs lists $5,400 for end-2027, Bank of America lists $4,750 for the first quarter of 2027, and the Reuters poll compilation lists $4,650 for 2027. Those three are not comparable with each other either: one is a year-end level, one is a quarter's average, and one is a poll.

Now set that spread beside the feed. The past-year high close was $5,318.40 on January 29, 2026, the most recent price is $3,929.70 as of November 3, 2026 at 16:50 UTC, and the gap to that past-year high is 25.9%. The feed also gives a past-year low of $3,898.90 on October 30, 2026. Across the last week gold moved −3.2%; across the last month, −5.2%; across the last year, −1.8%. In the feed's own words: Over the past month gold has fallen 5.2%, so the recent direction has been down. None of this tells you which target is closer. It tells you how much room the market has used in a year, which is the yardstick for how much a single published number deserves your attention.

Why targets get cut, and why they miss

Targets are revised because the world they were written in changes. The dated rows show that happening. The Goldman Sachs row comes from a page dated 2026-09-23 that describes the bank trimming its end-of-2026 fair value, while holding its end-2027 level. The Wells Fargo row comes from a page whose headline says the bank cut its 2026 target. A forecast is a snapshot of a view held on a date. When the price path or the inputs behind the view move, the snapshot is replaced, and the old one becomes a historical curiosity.

There are several honest reasons a published number misses.

First, the horizon is long and the market is jumpy. The feed's past-year high was $5,318.40 and its past-year low was $3,898.90, and a swing of that size over twelve months is typical of the metal, not a surprise. A forecaster who set a year-end level in the spring was predicting through a stretch that included that range.

Second, the inputs are themselves forecasts. A gold target usually rests on a view about demand from central banks, from funds, from jewelry buyers and from coin and bar buyers, plus a view on interest rates. If any one of those views is off, the target is off, and it is impossible to say from outside which input failed.

Third, history has plenty of sharp turns that nobody's target caught. London's closing price peaked at $850 an ounce on 21 January 1980, according to the Cleveland Fed's May 1980 commentary. That commentary also records the price falling to $650 two mornings later, and reaching a 1980 low of $480 on 17 March. Decades later, gold settled near $1,500 on 12 April 2013, about 22 percent below its record settlement of $1,920.30 in September 2011, as CNBC reported on the day. FTC testimony from 2014 describes the same stretch as gold rising from about $340 in 2003 to around $1,900 in September 2011, then closing 2013 around $1,200. We are not suggesting any of this repeats. We are showing that a sharp reversal is part of the record, and a target is a poor shield against one.

Fourth, publication itself lags. A dealer's compilation updated on 2026-11-01 can still list a bank's July figure. By the time you read any table, including ours, some rows are older than they look. That is why the table carries read dates, and why the review clock below exists.

What would have to be true for a target to hold

You cannot judge a target without knowing what it assumes. We did not read the bank notes, so we cannot list their assumptions. What we can do is name the kinds of conditions that sit behind any gold target and show what the World Gold Council's Q3 2026 data says about each, without claiming any link to the price.

Central banks. Official-sector purchases are the first thing to check: 321 tonnes in Q3 2026, per the Council's quarterly report, a figure its landing page sets 42% above a year earlier. A figure near the top of the table presupposes that this buying carries on. One near the bottom presupposes it fades.

Funds. Money moving through exchange-traded funds changes sign easily. The Council counts 211 tonnes of net additions in Q3 2026, following second-quarter outflows, and quotes record holdings of 4,256 tonnes. That swing makes funds a likely suspect whenever a forecast misses in either direction.

Retail demand. Jewelry buyers took 313 tonnes, down 17% on the year and the lowest third quarter the Council has on record, while coin and bar buyers took 268 tonnes, 19% fewer than a year before. A target that leans on strong retail appetite has to square with those readings or explain what changes them.

Interest rates. Many forecast notes lean on the path of rates. We cite the Federal Reserve only for the mechanism: the Federal Open Market Committee sets a target range for the federal funds rate, that rate is what banks pay to borrow reserves overnight, and changes in it are rapidly reflected in the rates lenders charge. We looked for World Gold Council wording on how rates, real yields or the dollar relate to gold and did not find any on the pages we read, so this page does not claim a link beyond the Fed mechanics.

Put plainly, a target is a bundle of bets. When the bundle is not shown, as it is not here, the number alone cannot be checked.

What this can't tell you

This page can't tell you whether gold is headed higher or lower, which published target will prove closest, or whether the price on your screen is a good one to pay. It cannot tell you that a bank still holds the view printed on its page, because we read secondhand reports and some of them were months old when we read them. It cannot tell you your own horizon, your budget or how much of a swing you can stomach, and those three decide more than any target does. For a way to think about your own decision, the decision framework on the home page walks through five questions. For the specific question of timing a purchase, see is now a good time to buy gold, and for the downside side of the story, will gold prices go down lays out what has pushed the metal lower before. If you want the other metal's equivalent list, it sits on the silver price prediction page, and the gold-silver ratio shows how the two prices compare. The forces that have moved both metals are gathered in will gold or silver go up, and if retirement accounts are on your mind, Augusta Precious Metals gold IRA explains the free information kit and the minimum purchase Augusta states.

Review clock

Targets read on November 3, 2026; reviewed every 90 days. Between reviews, a bank can change its view without this page knowing, so treat the date column as part of the figure.

Gold price prediction questions

Information, not financial advice. We do not tell you to buy or sell anything and we do not predict prices. The figures are reference numbers as of the time shown. For your own situation, ask a qualified professional.

Keep reading

• will gold prices go down: the drivers and past drops behind gold's downside risks
• the ratio between the two prices: how gold and silver compare, read from the live feed
• published silver targets: the same kind of table for the other metal