‘Gold is not done' and sovereign demand will drive price up to $4,900/oz in 2026 – Goldman Sachs' Dart

Gold has fallen sharply over the past four months, but Goldman Sachs’ Samantha Dart says the decline doesn’t signal the end of the 2026 rally. She expects central bank demand to push gold back toward the $5,000 level by year-end.

Analysis

Central-bank accumulation is a different demand function than ETF flows: it is slower, less price-sensitive, and more durable once reserve diversification is underway. That means the main market impact is not a straight-line squeeze higher, but a higher floor on downside and a lower realized-vol environment, which tends to favor physical proxies over beta-heavy miners. If the move resumes, royalty/streaming names should capture more of the upside per unit of risk than high-cost producers because their margin expansion is not diluted by labor, diesel, and sustaining-capex inflation.

The near-term catalyst is macro, not micro: softer real yields and a weaker dollar would give bullion the fastest path to follow-through over the next 2-6 weeks. The main reversal risk is a renewed rate-led dollar rally; if 10y TIPS yields keep rising, official-sector demand will likely only cushion gold rather than propel it, and equity multiples on miners can compress even if spot holds up. Over 6-18 months, the thesis gets stronger only if reserve diversification broadens beyond a handful of buyers and becomes visible in persistent import/flow data.

Consensus may be underestimating the gap between bullion and mining equities. Central-bank demand can keep spot firm while ETF outflows, jurisdictional risk, and input-cost inflation prevent a full re-rating of the equity complex, so chasing broad miners here looks lower quality than owning the metal. The cleaner expression is to own the asset with the most direct exposure to reserve-buying and avoid assuming that a higher gold price automatically fixes miners' capital allocation or valuation problems.

Key Decisions for Investors

• Buy GLD or IAU on any pullback over the next 1-3 weeks; use a thesis stop if 10y real yields resume a clear uptrend, since that would likely cap bullion despite official-sector demand.
• Pair trade: long GLD / short GDX for the next 1-3 months. This captures the view that physical demand supports spot more reliably than it re-rates miners; risk is a sharp ETF-inflow rebound into miners or a strong operational beat cycle.
• Prefer royalty/streaming exposure over producers: accumulate FNV or WPM versus NEM/AEM on weakness for a 6-12 month window. The trade benefits from gold upside with less operating cost inflation and lower geopolitical beta.
• Set an alert on DXY and 10y TIPS real yields rather than chasing price alone. If real yields fall back and GLD starts attracting sustained inflows, consider adding beta via GDX; if not, treat rallies as tactical rather than a structural breakout.