Gold: Central Banks Are Rewriting the Price Floor
Key Takeaways
- Central banks purchased 863 tonnes of gold in 2025 and are projected to buy 755 tonnes in 2026 — both well above the pre-2022 average of 400-500 tonnes.
- The buyer base is broadening: Malaysia bought gold for the first time since 2018, and South Korea is adding gold ETFs to reserves for the first time since 2013.
- Gold at $5,007 has rallied 70% from 2024 lows despite elevated real yields and a strong dollar — central bank demand has overridden traditional pricing models.
- J.P. Morgan targets $6,300 gold by year-end 2026, while even the most conservative forecast (Morgan Stanley at $4,400) represents only modest downside from current levels.
Gold trades at $5,007 per ounce, and the most important buyers in the market don't care what the price is. Central banks purchased 863 tonnes of gold in 2025 — the third consecutive year above 800 tonnes — and another 755 tonnes are projected for 2026. This isn't speculative demand. It's sovereign balance sheet restructuring — a trend explored in our central bank explainer — on a scale not seen since the 1960s.
The consensus narrative frames gold's move above $5,000 as a safe-haven trade driven by geopolitical fear. That misses what's actually happening. Price-insensitive sovereign buyers now represent a permanent "base load" of demand that has structurally shifted gold's equilibrium price higher. Even January 2026's modest 5 tonnes of net central bank purchases — far below the 27-tonne monthly average of 2025 — came alongside a broadening demand base, with Malaysia buying gold for the first time since 2018 and South Korea announcing plans to add gold ETFs to reserves for the first time since 2013.
Investors treating $5,000 as a ceiling should consider that central banks are treating it as a floor.
The Sovereign Buying Machine
Central bank gold demand has operated at a fundamentally different level since Russia's foreign reserves were frozen in 2022. Before that year, net purchases averaged 400-500 tonnes annually. Since 2022, the figure has run closer to 1,000 tonnes — a structural doubling that only partially corrected in 2025's 863-tonne total.
The World Gold Council's March 2026 data reveals January's 5-tonne net figure masked important shifts beneath the surface. Uzbekistan led with 9 tonnes, lifting reserves to 399 tonnes — gold now represents 86% of its total reserves, up from 57% in 2020. China added another tonne, extending its buying streak to 15 consecutive months and pushing gold to nearly 10% of total reserves. The Czech Republic and Indonesia each added 2 tonnes.
Russia was the largest seller at 9 tonnes — a notable reversal from the country whose reserve freeze catalysed the entire central bank buying wave. Bulgaria's 2-tonne reduction was mechanical: it transferred gold to the European Central Bank upon joining the eurozone on January 1, 2026.
The 2026 forecast of 755 tonnes looks like deceleration, but it's partly an arithmetic illusion. Higher gold prices mean central banks need fewer physical tonnes to reach the same percentage allocation target. At $5,000 per ounce, 755 tonnes represents roughly the same dollar commitment as 1,000 tonnes at $3,750.
The Broadening Demand Base
What matters more than the tonnage total is who's buying — and the pool keeps expanding.
Bank Negara Malaysia purchased 3 tonnes in January, its first gold acquisition since 2018. This lifted Malaysia's reserves to 42 tonnes, or 5% of total reserves. The Bank of Korea announced plans to incorporate overseas-listed physical gold ETFs into its foreign reserve portfolio starting Q1 2026, marking its first gold-related investment since 2013. Korea currently holds 104 tonnes of physical gold — roughly 4% of total reserves — placing it 41st globally.
These aren't marginal players making token purchases. When a central bank that hasn't touched gold in 5-13 years begins accumulating again, it signals a strategic reassessment of reserve composition driven by the same forces that pushed Poland to add 102 tonnes in 2025 alone.
The World Gold Council's 2025 survey found 95% of central banks expected global gold reserves to increase over the following 12 months. That near-unanimity is extraordinary. The motivation is consistent across regions: reducing concentration in major foreign currencies and sovereign debt. Gold's appeal as a non-sovereign, non-counterparty asset has become the dominant reserve management theme of the decade.
Macro Backdrop: Rates, Dollar, and Inflation
Gold's rally above $5,000 has persisted despite conditions that would historically have crushed the metal. The 10-year Treasury yield sits at 4.28% as of March 13. The Fed funds rate remains at 3.64%. The trade-weighted US dollar index reads 120.55 — elevated by historical standards.
Traditional models that price gold as a function of real yields and dollar strength have been unreliable since 2022. Gold has gained roughly 70% from its 2024 lows near $2,970 to today's $5,007 while real yields remained positive and the dollar held firm. The explanation isn't complicated: central bank demand has added a structural bid that didn't exist in prior cycles.
Inflation remains a secondary tailwind. US CPI hit 327.46 in February 2026, with annual inflation still running above the Fed's 2% target. But gold's move hasn't tracked inflation closely — it's outrun it by a wide margin. The metal is responding to reserve diversification flows, not consumer price indices.
Wall Street's Price Targets Tell the Story
The range of institutional forecasts for gold has widened dramatically, and every major bank has revised upward.
J.P. Morgan raised its year-end 2026 target to $6,300 per ounce from $5,055, citing sustained central bank and investor demand. Wells Fargo's Investment Institute targets $6,100-$6,300. UBS projects $5,400. Goldman Sachs sees $5,000 — which gold has already reached. Morgan Stanley, the most conservative of the group, forecasts $4,400.
The spread between the most bullish ($6,300) and most bearish ($4,400) forecasts is $1,900 — a nearly 40% range that reflects genuine uncertainty about how far sovereign buying can push prices. But the consensus direction is unanimous: higher.
One detail stands out. Central bank and investor demand is projected to average 585 tonnes per quarter in 2026. Unlike the volatile ETF-driven flows of the 2010s, this demand comes from price-insensitive buyers with multi-year allocation mandates. They don't panic-sell during corrections. They buy on dips. That creates a fundamentally different market structure than any prior gold cycle.
Portfolio Positioning Above $5,000
Gold at $5,000 changes the portfolio calculus. At this price, a 5% allocation in a $500,000 portfolio represents $25,000 — meaningful enough that the entry point matters even for long-term holders.
The case for maintaining or building gold exposure rests on structural demand rather than momentum. Central bank purchases represent a multi-year programme that won't reverse on a single FOMC meeting or jobs report. The broadening buyer base — Malaysia, Korea, and potentially others returning after long absences — suggests the allocation shift has further to run.
The risk is crowding. When every bank targets $5,000+ and every central bank survey shows 95% expecting higher reserves, some of that demand is already priced in. Gold's 50-day moving average of $5,041 sits just above the current price, suggesting the rally has stalled near-term even as the structural thesis remains intact.
For investors already holding gold, the data supports staying put. For those underweight, the case for adding on pullbacks toward $4,800-$4,900 is stronger than chasing at $5,000. The central bank bid provides a floor, but it doesn't eliminate volatility — January's price swings proved that.
Conclusion
Central bank gold buying has fundamentally altered the metal's pricing dynamics. The 863 tonnes purchased in 2025 and projected 755 tonnes for 2026 represent a permanent shift in demand that traditional macro models fail to capture. When price-insensitive sovereign buyers treat gold as essential infrastructure for reserve diversification, $5,000 becomes less of a milestone and more of a waypoint.
The broadening demand base — from established accumulators like China and Poland to returning buyers like Malaysia and Korea — signals that this isn't a trade. It's a structural reallocation that could persist for years. Gold's next move depends less on Fed policy or dollar dynamics than on how many more central banks decide their reserves are too concentrated in sovereign debt. Based on current survey data, the answer appears to be: most of them.
Frequently Asked Questions
Sources & References
www.jpmorgan.com
www.cmegroup.com
Disclaimer: This content is for informational purposes only and does not constitute financial advice. Consult qualified professionals before making investment decisions.