Gold's Structural Bid: Central Banks, De-Dollarization, and What Our Trade Flows Show
Executive Summary
Gold has entered a structural re-pricing regime. The metal’s move from the $1,800 range in early 2023 to sustained levels above $3,100 in 2026 is not a speculative overshoot — it reflects a fundamental shift in how sovereign balance sheets are being constructed. Central bank purchases have averaged over 1,000 tonnes annually for three consecutive years, a pace not seen since the Bretton Woods era. Meanwhile, dollar-denominated reserve assets as a share of global foreign exchange holdings have fallen to roughly 57%, down from 65% a decade ago.
This report draws on publicly available data — World Gold Council reserve statistics, IMF COFER disclosures, central bank balance sheet filings, and customs-level trade records — to examine the signals embedded in gold’s supply chain. From mine output and refinery throughput to port-level shipment data, the picture that emerges goes beyond the price chart. This report examines the structural forces underpinning gold demand and the investment implications for the quarters ahead.
Key findings:
- Central bank buying remains the dominant marginal demand driver, with China, Poland, India, and Turkey accounting for the bulk of official-sector accumulation since 2022.
- Physical gold is being rerouted through new intermediary hubs, notably the UAE and Singapore, reflecting geopolitical hedging behavior.
- Refinery utilization in Switzerland — the world’s processing bottleneck — has remained above 85% capacity for 18 months, signaling sustained physical tightness.
- De-dollarization is incremental but directional, and gold is the primary beneficiary among non-dollar reserve assets.
Central Bank Gold Accumulation: The New Normal
The scale of official-sector buying
The post-2022 acceleration in central bank gold purchases represents a regime change in reserve management. According to World Gold Council data, central banks added an estimated 1,037 tonnes in 2023, 1,045 tonnes in 2024, and approximately 980 tonnes in 2025. Early indications for Q1 2026 suggest the pace has not meaningfully slowed.
Central Bank Net Gold Purchases, 2018–2025 (tonnes)
To put this in context, annual mine supply stands at roughly 3,600-3,700 tonnes. Central banks are absorbing approximately 27% of newly mined gold — a structural demand floor that did not exist at this scale prior to 2022.
Who is buying, and why
The buyer composition matters as much as the aggregate numbers.
People’s Bank of China (PBoC): China has officially reported adding over 300 tonnes since late 2022, though market consensus — supported by discrepancies between Hong Kong export data and reported imports — suggests actual accumulation may be 1.5-2x the official figure. Gold now constitutes roughly 5.5% of China’s total reserves, up from 3.3% in 2020, but still well below the 60-70% gold share held by major Western central banks. The gap implies substantial room for continued buying.
National Bank of Poland: Poland added 130 tonnes in 2024 alone, making it the largest European buyer. This reflects a broader trend among Central and Eastern European nations seeking to bolster reserve credibility amid proximity to the Russia-Ukraine conflict.
Reserve Bank of India: India has been a consistent buyer, adding 50-70 tonnes annually, driven by a strategic decision to diversify away from U.S. Treasuries. India’s gold reserves now exceed 850 tonnes.
Central Bank of Turkey: Turkey’s purchases have been more erratic, oscillating between aggressive buying and occasional selling to support the lira, but the net trajectory since 2020 is firmly positive.
The common thread across these buyers is not a coordinated conspiracy against the dollar — it is a rational, portfolio-level reassessment of sovereign risk. The freezing of Russia’s $300 billion in foreign exchange reserves in 2022 was the catalytic event. It demonstrated that dollar- and euro-denominated reserves carry sanctions risk, transforming gold from a legacy relic into a geopolitical hedge.
De-Dollarization Dynamics and Gold’s Role
The slow erosion of dollar dominance
De-dollarization is frequently overstated in popular media and understated in institutional research. The reality lies between the extremes. The dollar remains the world’s dominant reserve currency, trade invoicing unit, and safe-haven asset. No credible challenger exists on any of these dimensions in the near term.
However, the directional trend is clear. The dollar’s share of allocated global foreign exchange reserves has declined from approximately 65% in 2016 to roughly 57% in 2025, per IMF COFER data. This decline has been absorbed not by a single alternative currency but by a basket of smaller currencies (the Australian dollar, Canadian dollar, South Korean won, and Chinese renminbi) — and by gold.
USD Share of Allocated Global FX Reserves, 2000–2025 (%)
Gold as the non-sovereign reserve asset
Gold occupies a unique position in the de-dollarization framework: it is the only reserve asset with no counterparty risk and no sovereign issuer. In a world where geopolitical alignment increasingly determines whether your reserves are accessible, this property has become more valuable.
Our analysis of trade settlement patterns corroborates this shift. KSINQ data shows a measurable increase in bilateral trade agreements that include gold-linked settlement mechanisms, particularly among BRICS-aligned nations. While these arrangements remain small relative to dollar-denominated trade, they represent a growing infrastructure for gold’s monetary role.
The critical point for investors is that de-dollarization does not require the dollar to collapse for gold to benefit. Even a gradual, multi-decade reallocation of 5-8 percentage points of global reserves from dollar assets into gold would imply cumulative demand of 8,000-12,000 tonnes — equivalent to 2-3 years of total mine supply absorbed by official-sector buyers alone.
Supply Chain and Trade Flow Signals: The KSINQ Perspective
What physical flows reveal that price alone cannot
Publicly available data — customs declarations, refinery throughput statistics, and logistics filings — surfaces physical signals that price charts alone cannot capture. In gold markets, where opacity is a feature rather than a bug, these publicly reported flows provide differentiated insight.
Signal 1: The UAE and Singapore as intermediary hubs
Over the past 18 months, we have tracked a significant increase in gold transshipment volumes through Dubai (DMCC) and Singapore. Gross gold imports into the UAE reached approximately 2,500 tonnes in 2025, up from 1,800 tonnes in 2023. A substantial portion of this gold is re-exported — to China, India, and increasingly to central banks in Southeast Asia that prefer not to source directly from London or New York.
This rerouting pattern is a physical manifestation of geopolitical hedging. Buyers are diversifying not only what they hold but where and how they acquire it. The implication is that traditional London Bullion Market Association (LBMA) flow data increasingly underrepresents total physical demand.
Signal 2: Swiss refinery utilization
Switzerland processes approximately 60-70% of the world’s gold, with four major refineries (Valcambi, PAMP, Argor-Heraeus, and Metalor) collectively handling over 2,500 tonnes annually. KSINQ tracks refinery input and output volumes through trade declarations.
Current utilization rates have been running at 85-90% of nameplate capacity for over 18 consecutive months. This is historically elevated — typical utilization runs 65-75%. The sustained tightness reflects robust demand for refined bars, particularly the 1-kilogram bars preferred by Asian buyers, as opposed to the 400-ounce Good Delivery bars that dominate Western institutional markets.
Signal 3: Mine-to-market lead times
Our tracking of mine output shipments from major producing regions — West Africa, Australia, and Latin America — shows average lead times from mine gate to refinery have extended by approximately 12-15 days compared to 2023 baselines. This elongation reflects both logistics capacity constraints and increased competition among refineries for feedstock. It is a secondary indicator of demand strength that will not appear in headline supply-demand balances.
Signal 4: Scrap flows remain subdued
Despite gold prices sitting well above historical averages, recycled gold supply has not surged as prior price models would predict. Our data shows scrap inflows to major refining centers are running roughly 10-15% below what the current price level historically implies. We attribute this to two factors: (1) consumer and institutional holders anticipate further price appreciation and are reluctant to sell, and (2) a portion of previously recycled gold has been absorbed into central bank stockpiles and removed from the secondary market.
Investment Implications and Risk Factors
The case for sustained strength
The structural forces described above — central bank accumulation, de-dollarization, and physical market tightness — are not cyclical. They are unlikely to reverse within a 12-24 month horizon. Our base case is that gold maintains a floor in the $2,900-3,000 range, with potential to test $3,400-3,500 if central bank buying accelerates further or if a U.S. fiscal event (debt ceiling crisis, credit downgrade) triggers another reserve diversification impulse.
Key supportive factors:
- Negative real rates in several major economies continue to reduce the opportunity cost of holding gold.
- Geopolitical risk premium remains elevated, with no resolution in sight for the Russia-Ukraine conflict, Taiwan Strait tensions, or Middle East instability.
- ETF flows have turned positive in 2026 after two years of net outflows, suggesting Western institutional investors are re-engaging.
Risk factors to monitor
No structural thesis is without risks. The primary threats to gold’s current trajectory include:
- A decisive Fed hawkish pivot that drives real rates meaningfully positive. If 10-year TIPS yields sustained above 2.5%, the opportunity cost calculus shifts against non-yielding assets.
- A geopolitical detente — particularly any thawing in U.S.-China relations — that reduces the urgency of reserve diversification.
- Central bank selling. While unlikely in aggregate, individual central banks (notably Turkey) have sold reserves during domestic currency crises. A coordinated or large-scale reversal would be bearish, though we assign this low probability.
- A stronger-than-expected dollar rally driven by relative economic outperformance or capital flight into U.S. assets, which would create headwinds for gold priced in dollars.
- Regulatory action against gold intermediaries in the UAE or Singapore could disrupt physical flow patterns, though it would likely redirect rather than reduce demand.
Conclusion
Gold’s current bid is not a trade — it is a structural reallocation. The combination of central bank accumulation at historically unprecedented scale, the slow but steady erosion of dollar reserve dominance, and physical market signals confirming sustained tightness points to a commodity that is being re-monetized in real time.
Synthesizing publicly available flow data, the physical picture is unambiguous. Gold is moving in larger volumes, through new channels, to new buyers, with longer lead times and tighter refinery capacity than at any point in the past two decades. The market is not oversupplied — it is being structurally absorbed.
For portfolio allocators, the implication is straightforward: gold deserves a strategic weight, not a tactical one. The question is no longer whether central banks will continue buying, but how much — and what that means for a market where above-ground supply grows at barely 1.5% per year. In a world of fiscal expansion, geopolitical fragmentation, and reserve asset re-evaluation, gold’s structural bid has room to run.
This report is published by KSINQ for informational purposes only and does not constitute investment advice. Data sources include publicly available data and independent analysis: World Gold Council, IMF COFER, and national central bank disclosures.