EU T+1 on 11 October 2027: A Settlement-Readiness Programme for the Middle Office

The EU moves to a T+1 settlement cycle on 11 October 2027. That is not a target to aim near; it is a cutover date, and the middle office is where the compression lands hardest.

Most of the T+1 conversation is being run by people selling migration services, and it tends to fixate on the settlement instruction itself. That is the wrong end of the pipe. The instruction is the output. What breaks under a shortened cycle is everything upstream of it that used to have a full extra business day to complete: allocation, affirmation, confirmation, matching, funding, and the exception handling that mops up whatever went wrong. The date is fixed by an amendment to the Central Securities Depositories Regulation that was published in the Official Journal in October 2025 and applies from 11 October 2027. The UK and Switzerland are aligning to the same day, so a firm operating across those markets faces one coordinated cutover, not a staggered one. This is a reading of the readiness programme that fixed date demands, written for the people who have to own the post-trade estate.

What actually gets compressed

Under T+2, a trade executed on Monday settles on Wednesday. The middle office has the whole of Tuesday to allocate a block, agree economics with the counterparty, confirm, resolve any mismatch, and release a matched instruction to the custodian or CSD. Under T+1, the whole of that day disappears. Trade date becomes the working day. The industry roadmap for the EU transition recommends that allocations and confirmations be completed as soon as possible and no later than 23:00 CET on trade date itself, using standardised electronic messaging rather than email and spreadsheets. Whether or not that specific time survives into the final rulebook, the direction is not in doubt: the affirmation and confirmation cycle that firms currently run overnight, or first thing the next morning, has to finish on the day of the trade.

The failure mode is predictable. Any process that today relies on a manual step, a batch that runs after close, a person in another time zone picking it up next morning, or a tolerant matching window, becomes a source of fails. And fails are not free. The CSDR settlement discipline regime attaches cash penalties to failed settlements, accruing daily until the trade settles. Halving the cycle mechanically raises the fail rate unless the upstream processes are re-engineered, which means the penalty line moves in the wrong direction on day one unless you have done the work.

Organise the programme around the date, not the function

The instinct is to hand each team its own T+1 workstream and let them optimise locally. That produces a set of individually improved processes that still do not chain together inside a single trade date. A better structure works backwards from the settlement start time and treats the day as a latency budget to be spent, not a series of independent tasks.

• Affirmation and allocation. Same-day affirmation is the single highest-leverage change. If blocks are allocated and economics affirmed automatically on trade date, most of the downstream compression solves itself. This is where the latency budget for automating allocations and confirmations has to be modelled explicitly, cut-off by cut-off.
• Static and reference data. A shorter cycle removes the slack that used to absorb a stale standing settlement instruction or a wrong place-of-settlement. Treating standing settlement instructions as governed master data, cleaned before cutover, prevents a whole class of avoidable fails.
• Funding and inventory. Cash and securities have to be in the right place a day earlier. This re-times FX, funding and securities lending recalls, and the misalignment between a T+1 securities leg and a T+2 currency leg is a real operational hazard for cross-currency trades.
• Exception handling. With no spare day, a break has to be seen and worked in near-real time. The exception queue stops being an overnight report and becomes a live control.

Reconciliation moves from overnight to intraday

This is the change most firms underweight. A reconciliation that runs once, overnight, against the prior day’s positions is a T+2 artefact. It tells you about a break the morning after there is time to fix it. Under T+1 the break has to surface while the trade date is still open. That means position, cash and instruction reconciliations running intraday, feeding an exception workflow that routes each break to someone who can act on it before the settlement window rather than after. The engineering work is not glamorous: event-driven reconciliation instead of batch, a break taxonomy that distinguishes a data mismatch from a genuine economic disagreement, and routing rules that get the right break to the right desk without a human triaging a flat list. None of it demos well. All of it decides whether you settle on day one.

Instrument the exception queue before you need it

You cannot manage a fail rate you cannot see. Before cutover, the exception queue needs to be measured: how many breaks per day, of what type, how long each takes to resolve, and which counterparties or instruments generate a disproportionate share. That baseline, captured under the current T+2 cycle, is what tells you where the compression will hurt. A counterparty that resolves affirmation breaks by lunchtime the next day is fine today and a problem the day the extra day disappears. Building that instrumentation is the subject of instrumenting the exception queue against CSDR penalties, and it is the part of the programme most likely to be left until testing, which is too late to change a counterparty’s behaviour.

Test against the cutover, not the average day

The US moved to T+1 in May 2024 without the disaster some predicted, and the lesson worth taking from it is that the firms that coped had rehearsed the busy day, not the quiet one. A quarterly options expiry, a large index rebalance, a public holiday in one leg of a cross-border trade: those are the days the compressed cycle is stress-tested, and they are exactly the days a lightly-rehearsed process falls over. Testing has to cover volume peaks and the awkward calendar cases, and it has to be sequenced early enough that a failed test can still change a process, a data feed or a counterparty arrangement before October 2027. Because a firm operating across the EU, UK and Switzerland faces one coordinated cutover across three rulebooks, the test plan has to prove all three settle cleanly on the same morning.

The date will not move for a firm that is not ready. It will simply arrive, and the unready firm will discover its middle office was the constraint all along, one penalty accrual at a time.