The Dutch Moved 14,000 Gold Bars To London. 1,163 Had Made The Trip Before.
De Nederlandsche Bank calls it crisis preparedness. The bar list says more than that: 1,163 bars in the shipment appear on the lists of eight earlier central bank moves, all ending at one London vault.
A months-long reporting project. Documents cited below are held in The Vault and available to readers.
AMSTERDAM, Netherlands — Of the 14,112 gold bars De Nederlandsche Bank shipped from Amsterdam to London last month, 1,163 have made the crossing before.
Not bars like them. Those bars: same serial, same refiner, same fineness, same gross weight to the hundredth of a gram, on the published list of an earlier central bank shipment. Nine of them have crossed three times.
The bank calls the move crisis preparedness. That is a true reason for this leg. It is not a reason for the rotation.
What was announced, and why it is checkable
On 2 September the bank confirmed it had moved roughly 176 tonnes of reserve gold to a London vault and described the move in a single sentence.
It also published a bar list, as it has done since 2014 and as most central banks now do. The list is the only reason this story exists. The most transparent participant in what follows is the one being written about, which is a poor incentive, and we would rather say so here than bury it in a footnote.
How the bars were matched
A bar list gives four things: a serial, a refiner, a fineness and a gross weight.
Matching on the refiner alone would be meaningless — four refineries cast most of the world's central bank gold, and a mark is not an identity. Matching on serials alone is nearly as bad, because serial ranges repeat across refiners and across decades.
All four fields together do not repeat. We matched the Amsterdam list against eight earlier repositioning shipments announced between 1968 and 2019, whose lists were either published at the time or obtained since.
Dr. Marijke Oostendorp-Baye, who does bullion logistics for insurers and had no contact with this story beyond her fee, ran the same match from the raw lists without seeing our working. She returned 1,163. We had returned 1,171. The eight-bar difference was ours, from a misread 1974 microfiche, and hers was right.
The rotation
Eight shipments, and one destination. The same London vault, every time.
And then, every time, a departure. After each arrival the matched bars fall off that vault's published holdings — median 41 days, range 27 to 96 — and reappear, between four months and eleven years later, in the reserves of the next central bank to announce a repositioning of its own.
Each bank describes its own leg in its own domestic language. A rationalisation of custody, in Ottawa in 1968. An audit convenience, in Bern in 1974. Liquidity, twice. Crisis preparedness, now.
Nine separate reasons, in five languages, for one loop.
Prof. Eamon Achterberg-Ruiz, who has spent twenty years on the history of bullion custody and accepts the match without reservation, thinks the loop is real and dull. "Institutions inherit habits from the clerks who set them up," he told us. "You are describing a queue. Somebody made the queue in 1949 and everybody has been standing in it politely ever since because leaving it would require a memo nobody wants to write." He may be right. The queue is still a queue.
The books that stop in 1991
Twenty-four volumes of minute books in this newspaper's vault record a shipping committee that met between 1949 and 1991 and numbered its work in legs.
A leg has an origin bank, a destination vault, a tonnage and a date. It does not have a purpose. It does not have a counterparty. The minutes call the recipient the account, thirty-one times, and never once expand it.
Five of our eight shipments are in the books, as legs 4, 9, 16, 22 and 31, and they match the public announcements to the day.
Leg 31 is the last entry. The books stop in February 1991, mid-volume, sixty-one blank pages before the end of the ruled paper. Three of our eight crossings happened after that, and we cannot tell you who scheduled them.
The second mark
1,163 is also the number of bars on the Amsterdam list that carry a second stamp.
It sits below the refiner's mark, struck rather than cast, and it is a single character. Our correspondent photographed it on forty-one bars directly. The bank's own published imagery carries it on the rest.
The two sets are the same set. Every bar that has crossed before has the mark. Every bar with the mark has crossed before. Neither we nor the refiners can tell you what it means; two of the four say the stamp is not theirs, and two have not replied.
What this desk takes from it
That the gold is not being stored. Storage does not have a median.
Every bank in this story published its own leg honestly, on its own calendar, in its own language, and answered our questions when we asked. Nothing was hidden. It was only never assembled, because assembling it means reading nine bar lists that nobody reads and a set of minute books that stops before the fax machine.
The nine lists, our matching code and the eight bars we got wrong are published with this story. If a reader can break the match, this newspaper will print that on the front page.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
We took nine bar lists — serial, refiner, fineness and gross weight to the hundredth of a gram — and matched them on all four fields together, then had the whole match run again from the raw lists by an analyst who had not seen our result. The 1949–1991 minute books have been in this newspaper's vault since 2019; the cross-reference against public repositioning announcements had never been run until this month.
Who we spoke to
- Dr. Marijke Oostendorp-Baye, Bullion logistics analyst, Rotterdam. Commissioned to reproduce the match from the raw lists, without sight of our result; paid at her standard commercial rate July–August 2026 Returned 1,163. We had 1,171. The eight-bar difference was our transcription error off a 1974 microfiche.
- Halvard Ntumba-Reisinger, Vault supervisor, London, 1974–2003 (retired). Interviewed in person over two days, with a third session by telephone 11–12 August 2026 Remembers the arrivals and the pallet tags. Says he was never told where a pallet went next and never asked.
- De Nederlandsche Bank, Press office and reserve management. Written questions submitted 3 August 2026; full written response received 24 August August 2026 Printed unedited under Right of Reply. The bank also asked us for copies of PX-1920.
- Prof. Eamon Achterberg-Ruiz, Historian of central bank custody, Geneva. Interviewed by video call twice 29 July and 18 August 2026 Accepts the match and thinks the rotation is a settlement habit rather than a plan. His explanation is printed at length in the body.
- The 1949–1991 minute books, PX-1920, twenty-four volumes. Deposited with this newspaper in 2019 by the executors of a retired shipping clerk; chain of custody published in the Vault 1949–1991
Documents
- PX-1920 — Minute books of an unnamed shipping committee, 1949–1991, 24 volumes accepted
What we could not confirm
- Who schedules the legs now. The books stop in February 1991. We have found no successor body, no successor numbering, and nobody in eleven months of asking who will say the words 'leg thirty-two'.
- What 'the account' is. The phrase appears thirty-one times in the minutes and is never once expanded into a name, a country or an institution. We are not going to guess in print.
- What the second mark means. We have it on 1,163 bars and our own photographs of forty-one of them. Two of the four refiners say the stamp is not theirs. The other two have not answered in five weeks. That is the end of what we know.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Continental WireDutch Central Bank Says London Gold Move Is 'Crisis Preparedness'
Ran the bank's sentence and the tonnage, accurately, inside ninety minutes.
Had that we did not
The announcement in full, and the correct tonnage, which two other outlets got wrong.
Left out
That the bank published a bar list on the same page as the sentence. The list is the story and it was sitting next to the quote.
-
The Hollow PostTHEY ARE MOVING THE GOLD BECAUSE THEY KNOW WHAT IS COMING
Read the shipment as the tell before a monetary event, dated to this autumn.
Had that we did not
That the same bars keep moving. They do, and this newspaper agrees.
Left out
That the rotation has been running since at least 1949 at roughly one leg every eighteen months, which is not a countdown. A thing that has happened thirty-one times is a practice.
-
They Buriedthis newspaperThe Dutch Moved 14,000 Gold Bars To London. 1,163 Had Made The Trip Before.
Stopped reading one bank's announcement and read nine banks' bar lists against each other.
Had that we did not
The eight earlier lists, our matching code, and the eight bars we got wrong before an outside analyst corrected us.
Left out
PX-1920 has been in our own vault since 2019. Nobody ran it against a public bar list until a wire item made the subject topical. The rotation was documented on our premises for seven years and we needed a news peg to look. — V. Ashcombe-Doyle, standards editor
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.