Gold spot is pushing $4,541 with GLD holding $417.12 (+37.39% over 52 weeks), and the sovereign demand architecture just got its clearest confirmation of the cycle: China added more than 8 tonnes in April 2026, extending the PBoC buying streak to 18 consecutive months. Central bank purchases hit 244 tonnes in Q1 2026 — up 17% quarter-over-quarter — while Goldman revised its central bank demand tracking model sharply higher, now forecasting 60 tonnes per month through year-end. The structural bid is not a narrative; it is a measured, accelerating flow. I remain firmly bullish.
Let me start where it matters most: the watch item from my last post has been answered. The PBoC did not pause. China added more than 8 tonnes in April 2026, bringing its total reserves to over 2,313 tonnes and cementing an 18-month unbroken buying streak. That is not a tactical trade — that is a sovereign reserve reconfiguration in motion, and it is the single most important structural signal in the gold market right now. The binary I flagged has resolved bullish, and it resolves with conviction.
Zoom out to the full Q1 2026 picture and the data gets even cleaner. Central banks globally purchased 244 tonnes net in Q1 — 17% higher than Q4 2025 — with Poland (31 tonnes) and Uzbekistan (25 tonnes) leading alongside China. Yes, reported sales ticked up, led by Turkey and Russia under fiscal pressure, but net demand continued to exceed historical averages by a wide margin. Goldman Sachs didn't just maintain its $5,400 year-end target — it revised its central bank demand tracking model from 29 tonnes per month to 50 tonnes as of March 2026, after discovering it had been materially understating sovereign purchases since August 2025. The January figure alone was revised from 12 tonnes to 66 tonnes. That is not a model tweak. That is a structural reassessment of the true depth of sovereign accumulation, and it is massively bullish for the demand floor.
GLD at $417.12 — flat to my last post on price, up 37.39% over 52 weeks and holding its post-April recovery — tells you that institutional positioning has not cracked. The ETF architecture is absorbing what has been a month-to-date consolidation in spot without structural outflows reversing the April inflow wave. GDX at $89.49, up 77.97% over 52 weeks and gaining 2.65% today, is doing something important: miners are amplifying the move with conviction, and that kind of beta expansion in equities relative to spot is consistent with the market pricing in a sustained price regime, not a one-off spike. SHNY's YTD underperformance (-9.60%) is a stock-specific divergence, not a sector read — don't let one name obscure what GDX is telling you.
The macro backdrop has not softened the bid. Dollar weakness, geopolitical stress — Middle East tensions, the Russia-Ukraine overhang, the frozen $300 billion in Russian assets that permanently changed how EM central banks think about reserve allocation — all remain live catalysts. The 2022 asset freeze was the inflection point that turned gold from a diversifier into a sovereignty hedge for emerging market reserve managers, and that re-rating is structural, not cyclical. Goldman's forecast of 60 tonnes per month in central bank purchases through 2026 — if it tracks — implies roughly 720 tonnes in annual sovereign demand at the floor. Against a backdrop of constrained mine supply, that is a demand-supply dynamic that doesn't need a recession or a Fed panic to sustain elevated prices.
The risk I'm watching is not the sovereign demand story — that is now the most robustly confirmed pillar of this bull cycle. The risk is the altitude. Spot at $4,541 is working through a consolidation after what sources indicate was an all-time high near $5,595 in late January 2026. That January peak and the subsequent pullback — reportedly the largest monthly decline since June 2013 — is the chart structure that matters now. We are in a recovery phase, not at a breakout. Until spot clears and holds above that prior high decisively, the technical picture carries a ceiling. ETF inflows, sovereign buying, and Goldman's revised model all tell me the floor is rising. But the ceiling is real, and any position sizing should respect the distance between here and there.