July 19, 2026 | Maybe Central Banks Are Gold-Price Sensitive After All

For a while now, analysts (including this one) have been predicting that central banks will keep buying gold regardless of its price, thereby smoothing out the normal ups and downs in that market.

But recent action implies that central banks do care about what they’re paying for gold — which, at this point in the cycle, is a good thing. Drawing on work from Goldman Sachs and Goldfix, here’s the story:

Central banks aggressively bought gold in the first few years of this decade, when it was hitting resistance at $2,000/oz (insanely cheap in retrospect). Then, when gold took off in 2024, the banks scaled back their buying.

Then — the important part for our purposes — when gold corrected in 2026, the central banks stepped up their buying.

This looks like bargain hunting, with China setting the pace:

The following bullet points are from Goldfix:

• Goldman Sachs estimates central banks bought 81 tonnes of gold in May, with its three-month seasonally adjusted pace rising to 67 tonnes per month, nearly four times the pre-2022 average of 17 tonnes.
• China was the largest identifiable buyer at an estimated 48 tonnes. Goldman says the renewed acceleration in official demand should help establish a floor beneath gold while hawkish Federal Reserve pricing pressures the market.
• Goldman maintains its $4,900 per ounce end-2026 forecast. Emerging-market reserve diversification remains the structural anchor, while the bank expects the current rates-related headwind to fade because its economists forecast no Fed hikes.
• Medium-term risks remain skewed higher. Private portfolios still hold relatively little gold, while geopolitical fragmentation and concerns over Western fiscal sustainability could broaden diversification beyond central banks.

Some Caveats

The outlook for inflation, growth, and interest rates is muddled by the Iran war. When the short-lived ceasefire took effect, oil prices plunged, and the June CPI fell by 0.4%, which made a Fed rate hike less likely.

When the shooting resumed, oil prices spiked, presumably pushing the next CPI reading back above the Fed’s target range.

The result: Near-term economic growth, inflation, and Fed behavior are a muddled, unpredictable mess, which might impact the gold price along with everything else.

Meanwhile, central bank gold buying is partially self-reported, and Chinese stats are notoriously self-serving and unreliable. So we shouldn’t put absolute trust in any given month’s numbers.

Think of central bank gold buying as one data point among many. Though right now, it’s a very positive data point.

Goldman Sachs estimates central banks bought 81 tonnes of gold in May, led by 48 tonnes of identifiable Chinese demand. The bank says official accumulation should provide a price floor during temporary Fed-related weakness, while reserve diversification, geopolitical fragmentation and low private allocations support its $4,900 end-2026 target.

John Rubino July 19th, 2026

Posted In: John Rubino Substack

Next: