Gold meta-analysis, part II
The framing has collapsed into one word: collateral.
Not investment advice. Claude can make mistakes.
Six weeks on, the April thesis ‘Is Gold money yet?’ got its confirmation and its asterisk in the same breath.
The flip went official
The single biggest development since April: what was Gromen’s contention is now in an ECB document.
On June 2, in its annual report on the international role of the euro, the ECB put gold at 27% of global official reserves by market value at end-2025, ahead of Treasuries at 22% and the euro at 15% (Nomi Prins’s read of it). In April, the reserve-asset flip was the thesis — Gromen’s “phase change, not a rally,” which he dated to roughly 1995. Six weeks later it’s the stated position of a G3 central bank.
One nuance worth keeping honest: the “first time since” date is measure-dependent. Gromen said since ~1995; the ECB framing gets read as since 1971. Those are different denominators — currency-mix of FX reserves versus total reserves at market value — so don’t over-index on the year. The direction is the durable part; the headline date is not.
The catch nobody’s leading with
Here’s what the gold accounts aren’t putting in the headline. The same ECB report attributes most of the share gain to valuation, not buying. Gold rose roughly 60% in 2025 after about 30% in 2024 (Prins); on price alone, its share of any reserve pool climbs without a single fresh ounce.
And net official buying actually slowed: 863 tons in 2025 — a figure that shows up verbatim in two separate June posts (Prins, TSCS) — against a prior-decade average near 470–473. Still well above trend, but down on 2024, and per the UBS-citing posts (Innes, VBL) concentrated in a handful of states (Turkey, Poland, Kazakhstan, China, India). TSCS, the most intellectually honest of the June bulls, makes exactly this its number-one falsifier: if the World Gold Council’s quarterly official-sector demand rolls over for two straight quarters, the floor under the trade goes.
So the six weeks delivered the April thesis’s biggest confirmation (the ECB) and its quietest caveat (it’s partly a mark-to-market artifact, and the marginal buyer is both slowing and concentrated) at the same time. Pillar #1 from April — “central banks accumulating at historic pace” — survives, but needs rewording: the stock revalued; the flow cooled.
And by mid-May, “cooled” was already too gentle. Ex-Goldman’s Jeffrey Currie (@CommodMkt, May 15) flagged that the tape had flipped from record buying to central-bank selling hitting headlines — citing BullionVault’s “Gold Erases 2026 Gains as Central Bank Selling Hits Headlines,” and a concrete case: the Turkish central bank selling roughly 120 tonnes (his figure) to defend the lira and fund energy imports, with analyst Bernard Dahdah noting in April that central banks are now selling gold to defend their currency and/or pay for energy.
Hold that against the list above: Turkey appears as a “structural buyer” in the reserve-diversification posts and as a “forced seller” in Currie’s — the exact buyer-to-seller flip TSCS built its mechanism on, and the exact falsifier it named, now showing up with a name and a tonnage attached. The bull floor isn’t gone, but it’s no longer hypothetical that it can crack. One honesty flag, reader to reader: this flip-to-selling read rests on a deliberately narrow base — one widely-shared X thread and one near-term liquidity call. Treat it as a development to watch, not a confirmed trend. If Turkey turns out idiosyncratic — one currency crisis rather than the leading edge of a pattern — this whole thread is noise.
What’s new on the board
1. Warsh is in the chair. In April the anxiety was Fed independence under the Trump–Powell standoff. By June that resolved into an actual new chair — confirmed 54–45, the narrowest margin in modern history (TSCS) — whose first meeting is June 17. TSCS makes that date the macro event of the year; the market prices ~99% no-change, which is precisely why the signal (tone, the dot plot) matters more than the decision. New, imminent, and absent from the April board.
2. The Winfree / Project 2025 thread (Grey Rabbit). Warsh has reportedly taken on Paul Winfree — author of the Project 2025 Fed chapter, which argues for commodity backing and, at its most radical, effectively abolishing the Fed — as a temporary contractor. The caveats are Grey Rabbit’s own: Winfree walked back the extremes in 2024, and Warsh has pledged the dual mandate. Suggestive, not evidence of intent. But it’s the first time a personnel link has been wired into the revaluation thesis.
3. A near-term bear with a real mechanism. Michael Howell (Capital Wars) is the June equivalent of April’s Defy the Odds, minus the short. His angle is liquidity, not rates or reserves: the PBoC has drained RMB 1.8tn (US$260bn) — roughly 4.5% — in 12 weeks since a March 2 peak, and gold in yuan looks set to break below RMB30,000/oz. It’s the only piece in the set pricing gold off Chinese monetary operations, and the only near-term caution with a named transmission channel. If he’s right, everyone else is watching the wrong dashboard for the next move.
The framing has collapsed into one word: collateral
April had six framings of the same facts (barbellization, phase change, credit-cycle deleveraging, debasement, capital-controls endgame, household rotation). The June set has largely converged on collateral:
VBL — the system is a contest over collateral, not currencies (”the collateral war”).
Miles Harris — “function, not fear”: gold’s first job was absorbing system distrust; its next is plumbing. He explicitly rejects the collapse narrative — “continuity, not collapse.”
Grey Rabbit — gold re-rated to make Treasuries sellable again; “gold becomes collateral for confidence.”
TSCS supplies the mechanism: rising real rates on a sovereign that can’t fund its own interest is the bullish setup, because the only exits are default or print, and both are gold-positive.
This is the real evolution, and it’s a trickle-down. In April, the collateral/revaluation idea lived with the heavyweights — Perera’s barbellization (gold as solvency layer beneath a still-liquid dollar) and Gromen’s statutory revaluation of the $42.22 book price to market, unlocking ~$700–800B. In June the same idea has arrived as retail-actionable product, and in the process it got inflated. Which is the bridge to the trade everyone’s asking about.
On Grey Rabbit’s “Called Shot” specifically
Separate the structure from the story, because they are not the same quality.
The structure is honest. A December 2026 $15,000 / $20,000 COMEX call spread: ~$300 risk, ~$500k max value per spread, sized at ~2% of book, framed explicitly as a lottery ticket you can afford to zero. Defined risk, transparent sizing, “if I’m wrong I lose 2%, my life doesn’t change.” No quarrel — that’s genuinely how you’d build a tail bet if you wanted one.
The story is where it strains. Three things to flag for any reader:
“Asymmetric” is not “positive expected value.” The spread is cheap (~$300) because the market assigns the outcome a tiny probability. The 1,000x payoff and the low cost are the same fact seen twice. To make money you need the true probability to exceed the implied one — and his case for that is “insiders are positioning,” which he concedes is unprovable (”I cannot prove that… Neither can anyone else”). An unfalsifiable edge is not an edge.
The “smoking gun” is the weakest evidence in the piece. Open-interest clustering at round-number strikes ($15k, $20k, $10k) is at least as consistent with cheap tail hedging and dealer positioning as with informed revaluation bets. Round numbers attract lottery flow — that’s a known retail/tail pattern, not a tell about insiders.
Grey Rabbit conflates two different revaluations. Gromen’s (April) is an accounting move: mark the $42.22 statutory price up to roughly today’s ~$4,500, unlocking ~$700–800B. Real, modest, plausible. Grey Rabbit’s requires the market price to first quadruple to $15–20k and then “force settlement” there — and his $4–5T windfall depends on it. Those are not the same claim, and his piece slides from the plausible one to the speculative one as if the first implied the second. Worth noting: his “absurd” $15–20k strikes are just Nieuwenhuijs’s $16,000 overshoot and Gromen’s $20,000 from April — except those were multi-year structural frameworks, and he’s stapled them to a six-month option with a June 11 entry deadline.
Net: it’s a defensible 2%-of-book tail bet if and only if you hold it as a near-zero-probability lottery ticket and resist the narrative’s pull to believe the odds are better than $300 implies. The risk isn’t the structure — it’s “we found the insiders” inflating the perceived probability above the price.
Where consensus sits now
More crowded, but the bears aren’t who you’d think. Correction to my own first draft: there is prominent pushback now — Jeffrey Currie (@CommodMkt) has been short gold since early March. The catch is that he’s a self-described perma-bull running a tactical short, with a path of “Gold 4,000 then 10,000.” So the secular-bear bench is still empty in both corpuses; what exists is a deepening bench of tactical shorts sitting on top of structurally-long views — Currie, April’s Defy the Odds, TSCS’s one-in-four flush, Howell’s liquidity dip. Even the bears here are bulls waiting to re-enter. That’s the honest shape of the “crowd”: not unanimous direction, but unanimous destination.
Same mechanism, opposite trade. The single most-cited bull mechanism is now also the bear’s entry. TSCS, Currie and Howell are all reading the same map — the marginal central bank flipping from structural buyer to forced seller to pay for energy, gold’s biggest bid vanishing. TSCS calls that flush a one-in-four dip to buy from the long side; Currie shorts it and goes back long once central banks turn dovish; Howell prices the near-term down on Chinese liquidity drainage. The disagreement isn’t about what happens — it’s purely which side of the same dip you stand on. When the bulls and the most credible short agree on the mechanism and differ only on the trade, the mechanism is the thing to watch, not the positioning.
Near-term is genuinely two-sided. Even the bulls concede the correction: gold peaked $5,595 on Jan 29, crashed (Grey Rabbit’s “11% on January 30”), bottomed near $4,098 in mid-March, and has gone quiet in a $4,400–4,800 range. TSCS puts a further near-term flush at one in four. Tavi Costa and Grey Rabbit lean the other way. Howell leans down. No edge in the tape.
The corpus itself skews more retail than April’s. The April cohort was anchored by data heavyweights — Nieuwenhuijs’s forensics, Gromen’s architecture, Perera’s structural work. This one skews toward paywalled, product-attached conviction pieces. That the story now arrives as a Dec-2026 option strike rather than a TIC-versus-gold reserve-accounting argument is, on its own, a late-ish-cycle tell: the thesis has finished travelling from the analysts to the retail tape.
What the Street says — the control group
Worth stepping outside the Substack cohort, because the sell-side is the quiet control here, and it cuts three ways.
Same side of the trade. Every bulge-bracket desk is structurally bullish. Goldman carries a year-end 2026 target around $5,400 — while explicitly rejecting a “commodity supercycle” and insisting gold be treated as a distinct monetary asset, not the opening act of a regime change. UBS is around $5,900–6,200 (with a $7,200 upside), JPMorgan the most aggressive near $6,300, and Deutsche Bank, Wells Fargo, BofA and UBP cluster at $6,000. So the cohort isn’t fringe — on direction, it’s aligned with Goldman and JPMorgan.
But the magnitude gap is the real story. Those Street targets sit roughly $5,000–6,300 — i.e. 0–30% above spot — and a Reuters poll of ~30 analysts puts the 2026 median near $4,750–4,900, essentially flat from here. Set that beside the cohort’s numbers: Nieuwenhuijs’s $8,000 base / $16,000 overshoot, Gromen’s and Grey Rabbit’s $20,000, Currie’s $10,000. Same direction, wildly different magnitude. The Street is pricing a continued grind; the cohort is pricing a phase change. If you only read Substack, you’d never learn that the professional median is “roughly unchanged.” And BofA’s Widmer adds a bull point the debasement crowd skips entirely — investor gold allocations are historically low, which is upside fuel precisely because most portfolios barely own it, the opposite of the cohort’s “everyone’s piling in” framing.
And the Street quietly guts the freshest thread. Pulled straight from the World Gold Council’s Q1 2026 release (the primary, not the headlines): central banks net bought 244 tonnes in Q1 — up 3% year-on-year and 17% on the prior quarter — even with the selling. Total reported official-sector sales came to 115 tonnes, comfortably outweighed by continued buying (Poland 31t to 582t, still chasing a 700t target; Uzbekistan 25t; the PBoC 7t to 2,313t), with the WGC noting unreported buying stayed elevated on top. On Turkey specifically — Currie’s whole case — the detail is the story: Ankara’s official holdings fell about 70 tonnes outright, plus roughly 80 tonnes via gold-for-FX swaps that, in the governor’s own words, return to reserves on maturity. Holdings then stabilised near 535t in April. The WGC’s verdict is one word — tactical — with explicit precedent in 2020 and 2023. So Currie observed something real, but a large slice of the “selling” is reversible swaps, the outright disposal was concentrated and brief, and in aggregate the official sector kept buying straight through it. The bear mechanism is sound in theory and, on the latest hard data, unconfirmed. The one place the cohort sounds most current — “central bank selling hits headlines” — is the one place the primary data is least on their side.
Bottom line
“Is gold money yet?” got a more official answer than I’d have bet on six weeks ago — the ECB now ranks it first by market value. But the same report says most of that was the price doing the work, and the marginal buyer slowed. The live tension isn’t bull versus bear — even Currie, the most credible short in either corpus, is a perma-bull who’ll be back long by $10,000, and the entire bulge-bracket Street sits on the same side, just two-to-four times more conservatively. It’s that the most confirmable version of the thesis (an accounting reserve flip, partly valuation-driven) and the most exciting version (a forced market revaluation to $15–20k) are being sold as though they’re the same story. They are not.
So: watch June 17; watch the WGC quarterly demand number TSCS flagged; and note that on the bear’s own chosen battleground — central-bank selling — the primary Q1 data shows the Turkey flush stalling (holdings stabilised, much of it reversible swaps) while the official sector net-bought straight through it. The bar for the forced-seller thesis is higher than the June posts make it sound. The thing that has actually happened is dull and large; the thing being sold is thrilling and unproven. Don’t confuse the two.
New names in this batch (the April source notes still stand)
TSCS (Strategist & Architect) — the standout. A fiscal-dominance / bond-market-first thesis built around June 17, with four dated, ranked falsifiers and its misses printed next to its hits. Highest rigor in the June set.
Nomi Prins (Prinsights) — clean reportage of the ECB milestone and the oil/reserve interplay; explicitly flags the valuation-vs-buying distinction. $6,000 year-end target, accumulation framing.
Michael Howell (Capital Wars) — the liquidity contrarian; the only near-term bear with a mechanism (PBoC drainage). Worth a permanent slot.
Jeffrey Currie (@CommodMkt) — a self-described perma-bull running a tactical short since early March (path: $4,000 then $10,000), on the Turkey-as-forced-seller thesis. The most credible “other side” in either corpus, and proof that the bear case here is a bull case with a timing overlay.
Egon von Greyerz; Metals and Miners — the conventional debasement lane (USD −99.24% since 1971; gold +11,119%). Conviction, not data.
VBL (GoldFix); Miles Harris — the “collateral / function” framing at its most articulate; Harris is the one writer here explicitly arguing continuity, not collapse.
Danny / Rob Kientz (CapitalCosm); Innes (Dark Side); Tavi Costa; QuantifiedStrategies — thematic and tactical fill, mostly downstream restatements of the reserve-flip architecture.
Full source list
The June 2026 corpus (16 posts, 3–5 June)
Quantified Strategies — Momentum Strategy for Stocks, Bonds, and Gold
Egon von Greyerz (VON GREYERZ) — GOLD WILL NOT GO UP IN THE NEXT 5-10 YEARS
Stephen Innes (The Dark Side Of The Boom) — Why Goldman Sachs and JPMorgan Think This Payrolls Report Could Rattle Markets
Danny / Rob Kientz (CapitalCosm) — Gold Just REPLACED U.S. Treasuries
Tavi Costa — A Sequence of Charts on Gold
TSCS (Strategist & Architect) — Gold Is Dead Money
Nomi Prins (Prinsights) — ECB: Gold Is Now the #1 Reserve Asset
VBL (GoldFix) — GoldFix PM
Metals and Miners — FIAT CURRENCIES ARE IN AN ETERNAL BEAR MARKET
VBL (GoldFix) — The Collateral War — vblgoldfix.substack.com/p/the-collateral-war
Metals and Miners — MONEY SUPPLY IS THE SIGNAL
Stephen Innes (The Dark Side Of The Boom) — UBS: Since 2022, Just a Handful of Countries Have Driven All Sovereign Gold Demand
VBL (GoldFix) — Special: Gold’s Hidden Reserve Windfall
Michael Howell (Capital Wars) — China And Gold
Miles Harris — The Longer Term Gold Outlook
Grey Rabbit Finance — The Called Shot
Added this round
Jeffrey Currie (@CommodMkt) — X thread, 15 May 2026 — x.com/CommodMkt/status/2055143663093662005.
Carried over from the April meta-analysis (”Is gold money yet?”, 26 Apr)
Luke Gromen — Forest for the Trees (FFTT) / Tree Rings. Founder/president of FFTT (est. 2014), ex-Cleveland Research and Midwest Research, CFA; macro-architecture node — fiscal dominance, de-dollarization, the “phase change, not rally” framing and the US Treasury gold-revaluation thesis. @LukeGromen.
Jan Nieuwenhuijs — The Gold Observer. Data-forensics node; triangulates IMF/WGC against London–Switzerland–SGE flows for unreported PBoC accumulation. $8,000 base / $16,000 overshoot framework.
Shanaka Anslem Perera — academic-style structural reserve work; “barbellization” framing (gold as solvency layer, dollar as liquidity layer).
Alasdair Macleod — MacleodFinance. COMEX plumbing, physical drains, Chinese household-flow specialist.
Lawrence Lepard — allocation/ETF-flow voice; conventional fiat-debasement lens.
Quoth the Raven (QTR) — Fringe Finance. Geopolitics and capital-controls lane.
Andy Schectman — via Adam Taggart’s Thoughtful Money; bullion-dealer ground-truth on delivery stress and repatriation.
Metals and Miners; StockTok; Defy the Odds — institutional-crisis, chart-plus-”faith in government,” and the lone skeptic (short since March), respectively.
Primary data
World Gold Council — Gold Demand Trends Q1 2026: Central Banks (29 Apr 2026) — https://www.gold.org/goldhub/research/gold-demand-trends/gold-demand-trends-q1-2026/central-banks
World Gold Council — Central banks resume net buying in April (Jun 2026) — https://www.gold.org/goldhub/gold-focus/2026/06/central-bank-gold-statistics-central-banks-resume-net-buying-april
European Central Bank — The international role of the euro annual report (2 Jun 2026), as reported by Prins — https://www.ecb.europa.eu/press/other-publications/ire/html/index.en.html
Street targets and price context (secondary aggregators — confirm at source before any figure anchors a chart or headline)
TheStreet (Goldman, UBS target revisions); GoldSilver.com (forecast roundup; “Why Turkey Sold Its Gold Reserves”); GBI Direct; GoldPriceTools; MEXC news (bank-target table); Visual Capitalist (2026 buyers/sellers); Holland Gold (Q1 selling update); ISA Bullion. Bank house-views cited: Goldman ~$5,400, UBS ~$5,900–6,200 (upside $7,200), JPMorgan ~$6,300, Deutsche Bank / Wells Fargo / BofA / UBP ~$6,000, Reuters poll median ~$4,750–4,900.
Visit MarketStack The Edit · MarketStack Terminal
MarketStack is free today. But if you value my work, you can pledge for a future subscription. MarketStack is an independent, anonymous publication summarising publicly available commentary and views from across financial media. Nothing here constitutes financial advice or a recommendation to buy, sell or hold any security. All views are a synthesis of public information. Past performance is not a guide to future results. This publication is not authorised or regulated by the Financial Conduct Authority. The author writes anonymously in a personal capacity.