Silver vs. gold in a reserve.

Public discussion of precious metals tends to collapse gold and silver into a single category — “buying metals,” “the metals,” “the precious complex.” The collapse is understandable; both have monetary history, both are mined and refined into bullion, both are quoted on the same exchanges. The collapse is also the most common error in this space.

Inside an institutional reserve framework, gold and silver are different instruments. They have different market sizes, different demand structures, different volatility profiles, different storage economics, and different treatment by the institutions that hold reserves at scale. Holding both at the same allocation rationale is a category mistake. Holding both for distinct, named jobs is reserve discipline. This note describes the differences and the architectural conclusion the Office takes from them.

The recurring annual demand-pattern figures from the WGC and the Silver Institute show two different shapes.

Gold demand is dominated by jewelry (which in most major-market data is itself a store-of-value behavior, not a consumption behavior in the strict sense), followed by central-bank reserve purchases, investment demand (bars, coins, ETF inflows), and a small industrial residual (electronics, dental, catalysis). Industrial demand for gold runs in the high single-digit percentage of the total. The dominant story in the gold demand chart is monetary.

Silver demand is dominated by industrial use: electronics, photovoltaics, electrical contacts, medical applications, solder, and a long tail of specialty uses. The Silver Institute’s 2024 World Silver Survey put industrial demand at over 50% of total fabrication demand — with photovoltaic (solar panel) demand alone now representing the single largest growth driver of the past five years. Investment, jewelry, silverware, and coinage make up the balance.

This is not a small distinction. Silver’s price has a real and substantial industrial-cycle component that gold’s does not. In a global manufacturing slowdown, silver demand softens in a way gold demand does not. In a manufacturing or solar-buildout cycle, silver demand pulls in a way gold demand does not. The two metals are more correlated than gold and equities, but materially less than they appear at a glance.

At household scale the volume-per-dollar argument that excludes silver from central-bank reserves loses most of its weight. A household is not vaulting thousands of tonnes; the marginal vault footprint of a 10–25% silver slice inside a private reserve is small in absolute terms. The structural case for a silver allocation inside a private reserve rests on two properties gold does not carry as cleanly: smaller exit lots and a different demand cycle.

Smaller exit lots: a 1,000-ozt silver Good Delivery bar represents roughly $35,000 at recent prices — against a 400-ozt gold Good Delivery bar at roughly $1.4 million. A household that wants to liquidate a portion of its reserve in measured tranches can do so in materially smaller dollar steps in silver than in gold, with corresponding flexibility for partial drawdowns or rebalancing actions.

Different demand cycle: silver’s industrial-demand share means its price is partly a function of global manufacturing and electrification cycles. That makes silver not a substitute for gold inside a reserve but a partial diversifier within the metals layer itself. Silver and gold are correlated — positively, in most regimes — but the correlation is not 1.0, and the residual carries the industrial-cycle signal that gold does not.

Neither of these properties is a forecast about silver’s direction. Both are structural — they describe the asset’s shape, not its expected return. The argument is not that silver will outperform; the argument is that a private reserve weighted entirely toward gold misses a job silver is well-suited to do at household scale.

The Office’s sample-brief composite carries a 75% gold / 20% silver / 5% platinum-group reference mix. It is not a prescription. Real briefs vary the mix according to the household’s objectives, the size of the reserve, the existing portfolio, and the exit pattern the household expects to need. A larger reserve with a longer horizon and no near-term drawdown plan can lean more heavily toward gold; a smaller reserve with a laddered drawdown plan or a manufacturing-cycle thesis can lean more toward silver.

The reasoning is the same in either direction: the metals are not interchangeable inside the reserve, and the brief explains which work each metal is doing rather than collapsing both into a single “precious metals” line item.