Central Banks Say They Are Still Buying Gold — and…

Gold price sits near $4,423 an ounce on Monday, up about 1.1% on the day and pressing back toward $4,400 after a week in which soft US data trimmed the odds of a Federal Reserve rate hike. Two things are worth writing about gold right now, and neither is a price call. The first is that the central banks buying gold say they intend to keep buying. The second is that the analysts paid to forecast where gold ends 2026 cannot agree, and the gap between their numbers is wide enough to be the story on its own.

That combination, a well-evidenced structural bid underneath a genuinely uncertain price path, is the honest state of the gold trade. FinanceFeeds covered the near-term rates trade when gold hit $4,414 before the July CPI print. This is the wider frame around it.

Where Gold Sits After CPI Week

The metal is coming off a week defined by the Fed. July inflation cooled to 3.4%, retail sales fell, and consumer sentiment softened, a run of data that pushed markets to price out an imminent rate increase. Because gold pays no yield, it gains most when the case for higher rates fades, so the soft data lifted it back toward $4,400.

Gold set an all-time high of $5,602 an ounce on January 29, then fell roughly 28% into the spring on hawkish Fed expectations, touching an intra-year floor near $4,000 before recovering. Even after that round trip it is up about 31% over the past year. The record and the retreat both matter here, because they are the reason the year-end forecasts scatter the way they do.

What the Central Bank Survey Found

The World Gold Council’s 2026 Central Bank Gold Reserves Survey, its ninth annual poll of reserve managers, found sentiment toward gold still climbing. According to the WGC, 89% of respondents expect global central bank gold reserves to rise over the next 12 months, and a record 45% expect their own institution’s holdings to increase, the highest that own-reserve figure has been in the survey’s history.

Reserve managers ranked gold’s performance in a crisis, its role as a long-term store of value, and its use as a portfolio diversifier as the top reasons to hold it, and a majority said they expect the US dollar’s share of global reserves to fall over the next five years while gold’s share rises.

Intent is not the same as flows, though, and the hard data complicates the survey. J.P. Morgan notes that central bank purchases averaged 225 tons per quarter from 2021 to 2025, roughly double the prior five-year pace, but that reported buying cooled sharply early this year: central banks sold 129 tons in the first quarter, led by Türkiye, leaving net reported purchases of just 16 tons.

Yet using London over-the-counter and Swiss refinery data, the WGC estimates actual first-quarter buying rose to 244 tons, with China a large unreported buyer. The survey says the intent is there. The flow data says the reporting hides more than it shows.

Investor Takeaway

Why the Year-End Forecasts Are So Far Apart

Here is where the honesty of the picture shows. The two most-cited gold price forecast targets for year-end do not just differ; they sit far above where gold trades now, and both have been cut as the metal fell.

J.P. Morgan, in its gold price forecast dated June 9, forecasts gold to push $6,000 an ounce by the end of 2026 and sees $6,300 as possible in 2027. That is itself a step down: the bank trimmed its fourth-quarter 2026 number from an earlier $6,300 and cut the nearer quarters more sharply, as gold’s spring slide undercut the earlier path. Goldman Sachs, back in January, raised its year-end target to $5,400 from $4,900, citing private-sector diversification on top of central bank and ETF demand, though that call predates the drop from $5,600 to $4,000 and should be read as a January view rather than a live one.

So the spread runs from Goldman’s $5,400 to J.P. Morgan’s $6,000, and both sit well above the current $4,423. Two of the largest research desks on Wall Street, looking at the same central bank demand and the same Fed, land more than $600 apart on year-end and each about $1,000 to $1,600 above spot. That distance is not a rounding error. It is a measure of how much the price depends on variables no survey can settle.

What the Fed Does to the Thesis

The single variable that most divides the forecasts is the Fed. Gold has no yield, so it struggles when rates are expected to rise and benefits when they are expected to fall. That is why the September meeting matters so much to the near-term gold price. As of August 17, CME FedWatch put the odds of a hold at about 67% and a hike at about 33%, down from nearly 50% before the soft data landed, with year-end hike odds still near 65%.

J.P. Morgan’s own bear case runs straight through that debate. The bank’s Greg Shearer identifies the biggest risk to the bullish view as a scenario where US growth and employment stay strong but inflation keeps accelerating, forcing the Fed to hike, which he says could crack investor demand and trigger sustained ETF outflows. He calls it a high bar for the year, but not an impossible one, and energy prices are the reason it lingers. It is the same tension FinanceFeeds flagged when the CPI print did not settle the September Fed debate. The structural bid sets the floor; the Fed sets the direction from here.

What to Watch

Three things decide which gold price forecast looks right by December. The first is the Fed path, because a hold keeps gold’s rate-relief bid intact while a hike would pressure it directly, exactly as J.P. Morgan’s bear case describes. The second is whether the reported central bank buying reaccelerates or the gap between reported and estimated purchases keeps widening, which will show whether the survey’s intent translates into flows, a question FinanceFeeds has tracked through China’s continued gold accumulation.

The third is the forecast spread itself: if Goldman and J.P. Morgan converge as the year runs down, the price path is clarifying, and if they stay $600 apart, the uncertainty the current tape reflects is real. The gold price near $4,400 is neither the record nor the floor. It is the market waiting to see which of these resolves first.

Investor Takeaway