Why Central Banks Keep Buying Gold While Institutions Hold Zero

The people who manage sovereign balance sheets, the most sophisticated institutional actors in the world, have been buying gold at roughly twice their historical pace for three straight years. The average private investor holds essentially none.

That asymmetry is the story. It is not a gold-price story, and it is not a forecast about where the metal goes next.

Central banks bought 1,092 tonnes of gold in 2024 and 863 tonnes in 2025, both figures far above the 2010-2021 annual average of roughly 473 tonnes. They kept buying even as gold set record after record, which means elevated prices did not deter the official sector at all. That is the anomaly worth explaining.

Here is the question your own portfolio is quietly asking: does the behaviour of the world’s reserve managers count as a meaningful signal for how you should think about gold, or is it noise you can safely ignore? By the time you finish reading, you will have the data and the framework to answer that for yourself.

The scale of official-sector accumulation that most private investors have missed

Start with the baseline, because the magnitude only registers when you see what changed. Across 2010 to 2021, central banks bought an average of about 473 tonnes of gold a year. Steady, unremarkable, background noise in the reserve-management world.

Then 2022 happened, and the baseline doubled. From that year through 2024, official-sector buying ran above 1,000 tonnes annually, peaking at 1,092.4 tonnes in 2024, according to the World Gold Council’s Gold Demand Trends report published on 29 January 2026.

In 2025, buying eased to 863.3 tonnes, a 21% year-on-year decline. On paper that looks like a retreat. It is not. That figure still sits at nearly twice the pre-2022 average, and the WGC forecasts roughly 850 tonnes for 2026, according to Reuters’ summary of the same dataset. The pace is being revised down modestly, not unwound.

Q2 2026 official-sector flows showed continued structural commitment, with a record 289 tonnes acquired in a single quarter, a pace that reinforces the WGC forecast of roughly 850 tonnes for the full year rather than a return toward pre-2022 norms.

The share data sharpens the picture further. The European Central Bank noted in research published on 11 June 2025 that central banks accounted for more than 20% of global gold demand in 2024, compared with around 10% across the 2010s. The official sector effectively doubled its footprint in the gold market.

What makes this hard to write off as a regional quirk is the breadth of the buyers. This is not one nervous government hedging a local problem.

For 2024, the leading purchasers were:

• Poland: 90 tonnes (top buyer)
• Turkey: 75 tonnes
• India: 73 tonnes
• China: 44 tonnes

For 2025, the standout names were:

• Poland: 102 tonnes, lifting total reserves to 550 tonnes
• Kazakhstan: 57 tonnes
• Brazil: 43 tonnes

Europe, Asia, and South America, all accumulating simultaneously. This is a structural shift in how sovereign institutions value gold as a reserve asset, not a trading signal you can time. And it appears set to continue.

89% of reserve managers surveyed by the World Gold Council in 2026 expect global central-bank gold holdings to increase over the next 12 months.

The question this leaves you with is simple. Do you treat a near-doubling of official-sector demand, sustained across three years and multiple continents, as information or as background noise?

What is driving sovereign institutions to gold, and why now

To read the signal, you have to understand the motive. And the motives stack in ascending order of weight: the deeper you go, the harder they are to reverse.

The sanctions trigger

The proximate cause is geopolitical, and the timing is not subtle. The surge in official gold demand maps directly onto 2022, the year of Russia’s full-scale invasion of Ukraine and the subsequent freezing of Russian foreign-exchange reserves.

ECB economists documented this directly in their June 2025 research, noting that gold demand for monetary reserves surged sharply after 2022 and has stayed elevated since. The logic is straightforward: a reserve held in another country’s bonds or bank deposits can be frozen with a signature. Gold in your own vault cannot.

Gold is not subject to default risk or to unilateral freezing in the way that foreign-currency bank deposits and sovereign bonds are, according to ECB research (June 2025).

De-dollarisation as a structural driver

The medium-term driver runs deeper than any single conflict. Central banks holding large stocks of US Treasuries and other Western assets have watched what sanctions can do, and they are diversifying away from dollar concentration.

Amundi’s October 2025 analysis characterised this as structural demand from emerging-market central banks seeking to reduce reliance on US dollar assets and shore up monetary credibility, not tactical trading. Gold is the politically neutral reserve, beholden to no single government’s debt.

The de-dollarisation trends reshaping global reserve management extend well beyond gold accumulation; shifts in trade-settlement currency choice, bilateral swap lines, and the composition of sovereign wealth funds are all moving in the same direction simultaneously.

You can see the shift in where the gold physically sits. An Investing.com analysis from 7 September 2026 noted a rising trend toward storing reserves in domestic vaults rather than with foreign custodians. That tells you something the volume data alone cannot: this is a decision about trust in the international monetary system, not an allocation to be traded in and out of.

Currency debasement and the bond market

The deepest motive is the one that will not reverse when the next ceasefire is signed. VON GREYERZ partners Jonny Haycock and Matthew Piepenburg, writing in September 2026, framed the accumulation against a backdrop of mounting fiscal pressures: sovereign bond markets under stress, government interventions in financial markets, inflation proving stickier than official narratives allow, and a speculative technology cycle showing signs of exhaustion.

Their central framing is that gold is being bought to preserve purchasing power amid routine currency debasement, not for speculative gain. That is a bet against the credibility of fiat money and the health of the bond market itself.

Here is why the motive matters more than the volume. If central banks are hedging against currency debasement and the weaponisation of dollar-denominated reserves, you face the same underlying exposures, just through different channels in your own portfolio.

The private investor gap: what the allocation data actually shows

Now hold that behaviour against what private portfolios actually contain. The single most striking figure is this: sophisticated long-term US institutional investors average roughly 0.2% of their portfolios in gold, according to a Kitco analysis published on 21 September 2026.

Not 2%. Not even half a percent. Twenty basis points.

Contrast that with the minority who do use gold meaningfully. A biannual survey by Coalition Greenwich and the World Gold Council, covering more than 400 institutions and cited by First Eagle Investments on 16 September 2026, found that among institutions holding gold, the average allocation is around 4%, and most treat it as a strategic position they expect to maintain or grow.

The more revealing number is participation. Only about 15% of surveyed institutions hold any gold at all. The other 85% hold zero explicit exposure. This is not a scatter of individual oversights; it is a market-wide structural pattern.

The structural reasons for low gold ownership among institutions run deeper than mandate restrictions alone; historical asset-class categorisation, benchmark construction, and the absence of a yield stream have collectively embedded a near-zero default allocation across decades of modern portfolio theory.

What makes the gap so stark is when it persisted. Gold averaged US$3,431.5/oz across 2025, up 44% year-on-year, and set 53 new all-time highs in the year, per the WGC. Total investment demand hit 2,175 tonnes, up 84%. Private institutions stayed on the sidelines through one of the strongest gold years on record.

Before concluding that 85% of institutions are simply wrong, the counterarguments deserve honest treatment:

• Opportunity cost: Gold pays no yield. In a high-real-rate environment, holding it means forgoing income from bonds and cash.
• Mandate and regulatory constraints: Many institutions cannot hold gold, or face capital treatment that makes it inefficient.
• Price-level risk: Buying after a 44% run-up risks entering at a peak. Reuters noted that record prices are expected to temper both central-bank buying and jewellery demand.
• Deflation scenario: In a deep deflationary shock, long-duration bonds and cash have historically outperformed gold.

There is a colder reading of the 0.2% figure, too. At that weighting, gold could double again from its 2025 average and the typical institutional portfolio would barely register it. The question is whether that insulation reflects considered wisdom or simple inertia. The data cannot tell you which; it can only tell you the gap is real and measurable.

Reading the signal: what central bank behaviour can and cannot tell private investors

So you have a measurable gap between the most institutionally sophisticated buyers on earth and everyone else. The temptation is to resolve it cleanly in one direction. Resist that.

Why the signal is harder to dismiss than usual

Central-bank buying is not a momentum trade or a short-term view. VON GREYERZ describes the accumulation as intentional, systematic, and indicative of a broader monetary-strategy shift, framing gold as a purchasing-power preservation asset rather than a speculative one.

The WGC’s institutional portfolio research from October 2022 points the same way: the institutions that do hold gold treat it as a strategic allocation they intend to keep or increase, not a position they trade around.

Central banks are not buying gold because they expect to profit from it. They are buying it because they no longer fully trust the alternatives, in the framing of VON GREYERZ analysts (September 2026).

That distinction is what gives the signal its weight. When the people who manage the world’s reserves quietly move toward the one asset that cannot be frozen or inflated away, it is worth asking what they are hedging against.

Where the analogy between central banks and private investors breaks down

Here is the limit. What a central bank hedges and what you hedge are not the same thing.

A central bank is defending against currency debasement, sanctions vulnerability, and fiat credibility risk at a national scale. Your hedging needs depend on your own liability structure, income requirements, and time horizon, none of which a reserve manager shares.

The historical risk-return data complicates any urge to match official-sector conviction with a large personal allocation. Quantpedia’s analysis suggested the average gold-holding institution carries roughly 1.7% exposure and that performance benefits plateau beyond low-single-digit allocations, though this figure is drawn from a single source and not independently confirmed.

There is a concentration caveat as well. With central banks now representing more than 20% of global gold demand versus 10% in the 2010s, pricing has become more dependent on official-sector flows. If that buying moderates, the marginal support underneath the price moderates with it. The signal from central-bank behaviour is real, but so are the reasons a thoughtful investor might read it differently.

Where this leaves gold’s role in a private portfolio today

The core argument is now visible. The gap between official-sector conviction and private-sector holdings is real, measurable, and has survived a year of record prices, which suggests it reflects structural underweighting rather than a temporary quirk.

That does not translate into a buy instruction. It translates into a set of conditions worth auditing before you decide anything. The central-bank signal becomes more relevant to your portfolio the more your own exposures resemble the ones reserve managers are hedging.

Work through these:

1. Dollar-asset concentration: How much of your portfolio sits in a single currency or in Western sovereign debt?
2. The real-yield environment: Gold’s opportunity cost is lowest when real yields are low or negative.
3. Existing inflation hedges: Do you already hold assets that address currency-debasement risk?
4. Time horizon: Gold’s case strengthens over longer holding periods, where currency risk compounds.
5. Mandate or account constraints: Are there structural reasons gold does not fit, as with many institutions?

89% of the world’s reserve managers expect to increase gold holdings over the next 12 months, while the average private institution holds 0.2%.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions. Past performance does not guarantee future results, and financial projections are subject to market conditions and various risk factors.

The gap between who buys gold and why they buy it narrows what private investors can ignore

Pull the two threads together and the asymmetry is plain. Central banks are buying at roughly twice their historical pace and expect to keep going; most private institutions hold zero. In 2025 alone, the official sector added 863 tonnes while the typical institutional portfolio sat at 0.2%.

The interesting part is not that private investors disagree with central banks about gold. It is that most have never formed a view at all.

That absence of a view is itself a position. Holding no gold implicitly bets against the systemic risks, currency debasement, fiat credibility, dollar concentration, that the world’s reserve managers have spent three years hedging against. It is a real bet, whether or not it was made deliberately.

The useful action here is not necessarily buying gold. It is deciding, consciously, whether your current exposure, including zero, reflects a considered view or an oversight. Audit whether the risks central banks are hedging exist in your portfolio in any form, and whether what you already hold addresses them. Then hold your position on purpose.

For investors who have worked through the five-point audit and concluded that some gold exposure is warranted, our dedicated guide to sizing a personal gold allocation walks through a framework for translating that conclusion into a specific portfolio weight based on existing currency exposure, time horizon, and risk tolerance.