A delayed payments release is fine. A delay without a new date is not
The Bank of England has pushed back its November 2026 RTGS standards release. Slipping a date in shared infrastructure is often the right call — but only if the replacement date and the reason land with it.
The Bank of England runs the system where banks hold money and settle large payments with each other. From time to time it updates the format of the electronic messages used to instruct those payments, and everyone has to switch on the same day. One such update was scheduled for November 2026. The Bank has now said it will be delayed.
Delaying a payments standards update is usually the right call — a broken release is far worse than a late one. But a delay is only half an announcement. The new date and the reason are the other half.
The most expensive word in financial plumbing is “delayed”
It does not move a single interest rate expectation. It will not lead a bulletin. But when the Bank of England said in late August that the November 2026 standards release for its real-time gross settlement service would be delayed, that sentence was read very carefully in several hundred change-management offices — and it rearranged a great deal of work that had already been paid for.
What is actually being moved
Real-time gross settlement, usually shortened to RTGS, is the account system at the central bank where the commercial banks hold balances and settle large payments between themselves one at a time, with finality. It is the bottom of the stack. Everything above it — card payments, salaries, house purchases, the funding leg of a securities trade — eventually resolves into an entry in that ledger.
A standards release is a change to the format and data content of the instructions sent into that system. Not the plumbing itself, but the language spoken through it. And the defining property of a shared language is that everyone has to switch on the same day. A standard adopted by half the participants is not a standard; it is an incident. That is why these releases are scheduled years out, announced publicly, and treated by every participant as a fixed point around which other projects are arranged.
The date is the product
This is the thing that is easy to miss from outside. In infrastructure change, the value delivered by the central bank is not only the technical specification. It is the coordination. Publishing a date is what allows a few hundred independent organisations, with different budgets, vendors and internal politics, to commit resource to the same change at the same time without negotiating with each other.
Which means a delay is not a gift of extra time. Delivery teams do not get eight months of leisurely polish. Contractors booked for a cutover window are either paid to wait or released and re-hired at a premium. Test environments that were configured to a spec sit frozen. Downstream work that was sequenced to start after the release either starts anyway on assumptions that may not hold, or slips in turn. The cost of a slipped standards release is not zero and it is not small; it is simply invisible, because it is spread thinly across balance sheets that are not the central bank’s.
The asymmetry nobody can legislate away
Inside any operator of critical infrastructure, the incentives point one way. A late release is embarrassing for a quarter. A broken release in the settlement layer is a national event, a parliamentary appearance and a decade of citation in other people’s risk papers. Faced with genuine doubt about readiness, any competent operator slips the date. They should.
But notice what that means. The party making the call carries the reputational downside of shipping badly and almost none of the financial downside of shipping late. When one side of a decision holds all the costs and the other holds all the discretion, you do not fix it by asking for better judgement. You fix it by making the record visible — so that the pattern of slippage, if there is one, is a fact rather than an impression traded between vendors.
The strongest case against me
Here is the honest version of the other side. Standards releases in payment systems are frequently delayed at the request of the industry itself. Participants who are behind schedule lobby for more time; the operator, weighing the risk that a minority arrives unready, grants it — and then absorbs the public criticism for a slip that its own users asked for. On that account, demanding louder accountability for delays punishes the one actor behaving responsibly and rewards the ones who could not deliver.
I accept the mechanism entirely. I draw the opposite conclusion. If the industry asked, say the industry asked. A published reason is what converts an anonymous slip into an attributable one, and attribution is the only thing that changes behaviour among participants who currently face no consequence for being the laggard that moved everyone else’s date. Silence about causes protects the unready, not the operator.
The boring layer under the exciting one
This is not a British peculiarity. Days after the Bank’s statement, the Reserve Bank of Australia opened a consultation on the role of its own settlement system in a tokenised financial ecosystem, citing efficiency, functionality and resilience in wholesale markets. Every central bank now has an ambitious settlement agenda of that kind.
All of it rests on the unglamorous layer: message formats, release windows, cutover weekends. An institution’s credibility on tokenised wholesale settlement is not established by a consultation paper. It is established by whether the mundane standards release it promised for November arrived in November — and, when it did not, by whether the institution said clearly when it would instead.
The Bank of England should ensure that this delay is accompanied — now, not at the next scheduled update — by a firm replacement date and a plain statement of what was not ready. Beyond this instance, it should maintain a public register of RTGS release dates as originally announced against dates actually delivered, updated at every change and carried forward across releases.
That register costs a spreadsheet and a page on a website. It would give participants, vendors and the Bank's own oversight committees a shared, checkable record of delivery performance in the layer that everything else in British finance sits on top of.
What would change this view
If the Bank's 27 August statement already contains a firm replacement date for the release and a stated reason for the move, and the release then lands on that revised date, my complaint has no object and the process worked as it should. Equally, if the Bank publishes a running register of RTGS release dates as originally announced against dates delivered, the transparency I am asking for already exists and I am late to it.
If either is true, the argument narrows to something much smaller: that the rest of the industry should be able to see the same record without hunting for it. If neither is true by the time the revised release lands, the case for a standing, published delivery record gets stronger, not weaker.
Worked from
- Delay to the November 2026 RTGS standards release
Bank of England · 2026-08-27