Central banks are buying more gold than expected, and purchases will increase further through 2026 – Goldman Sachs

Goldman Sachs says central bank gold purchases have been stronger than previously estimated so far in 2026 and now expects sovereign demand to rise further in the second half of the year. The firm revised its central bank gold demand model to account for gaps in official trade data, implying higher-than-assumed official-sector buying. The update is supportive for gold sentiment, though it is primarily an analytical revision rather than an immediate market catalyst.

Analysis

The most important second-order effect is not the absolute level of central-bank buying, but the implied tightening of float in a market where marginal supply is already price-inelastic. If official demand is running above prior models, the market loses a key source of latent sell pressure just as macro volatility and reserve diversification keep bid support intact; that combination tends to steepen rallies because there is less hedging inventory to absorb flows.

The biggest beneficiaries are the upstream and high-beta expressions on gold rather than the metal itself: producers with unhedged production, royalty/streaming names, and basket ETFs with operating leverage should outperform if the demand revision persists into H2. By contrast, downstream fabricators and jewelry demand are the most vulnerable to any further price acceleration, since their margin sensitivity is immediate and they do not get the reserve-management bid that sovereign buyers provide.

The catalyst path is asymmetric over the next 1-3 months. Near term, the market can re-rate on the idea that official sector demand has been undercounted; over 6-12 months, the key reversal risk is a stronger USD or a real-rate backup that forces reserve managers to pace purchases rather than abandon them. The contrarian read is that consensus may be overfocusing on “gold bullish” and underappreciating that the real trade is a volatility regime shift: when reserve buyers become more active, options-implied skew in miners can cheapen relative to realized upside, creating cleaner convexity than spot.

For GS, the earnings impact is limited, but the franchise benefit is broader: models that improve narrative accuracy on a sensitive macro commodity can reinforce trading and commodities client flow. If this update is perceived as another example of GS having better read-through on sovereign behavior, it modestly improves the firm’s positioning as a policy-aware commodities advisor, even if the direct P&L effect is small.

Key Decisions for Investors

• Long GDX vs short broad equities for 1-3 months: gold miners should capture operating leverage if sovereign buying keeps tightening the free float; use a ~5-7% stop if real yields spike.
• Buy unhedged producer basket on pullbacks (e.g., NEM, AEM) for a 3-6 month window: asymmetric upside from higher realized prices with limited incremental cost inflation; target 15-20% upside if gold continues to grind higher.
• Pair trade: long royalty/streaming names (FNV, WPM) vs short jewelry-exposed names or consumer discretionary importers for 2-4 months, as rising gold prices are more painful downstream than upstream.
• Use call spreads on GLD or GDX rather than outright longs: the signal is supportive but not explosive, so defined-risk convexity is better than paying for downside carry in case the USD rallies.
• For GS specifically, stay neutral-to-slightly-long into any commodities franchise update, but avoid chasing; the direct EPS impact is modest, so the trade is on improved client activity and not a standalone earnings re-rate.