FX and Funding Under T+1: Re-Timing Your Cash and Securities Lending

A settlement cycle does not just move the date the securities have to be there. It moves the deadline for the cash and the stock that back them — and under T+1 those deadlines fall on the previous day, which for a firm funding in another currency can mean the previous evening.

The EU moves to a T+1 settlement cycle on 11 October 2027, following the amendment to the Central Securities Depositories Regulation and ESMA’s recommendation of that date. The UK and Switzerland are aligning on the same day. Most of the programme attention has gone to allocations, confirmations and the middle-office timeline, which is where it should go. But the part that quietly breaks first is funding: the foreign exchange that pays for the securities, and the stock on loan that has to come back before it can be delivered. Both ran comfortably under T+2. Under T+1 the slack disappears, and the failure mode is not a system error — it is cash or stock that arrives correct but late.

The extra day was a funding buffer, and it is being removed

Consider a European fund buying US equities, or a US fund buying European securities. The purchase settles in one currency; the fund holds another. Somewhere between the trade and the settlement, an FX trade has to convert the cash and that cash has to be where the custodian needs it. Under T+2 there was a full business day to execute the FX, net it, submit it for payment-versus-payment settlement and still have the currency in place. That day was doing real work. It absorbed time-zone gaps, late allocations, a missed cut-off, a broken instruction — all the ordinary friction of moving money across borders.

T+1 removes it. The FX now has to be executed and funded on trade date or early on the settlement date itself, and for a firm operating several time zones away from the settlement currency, “early on the settlement date” can fall outside its own working hours. The trade has not become harder. The window in which everything has to be right the first time has become much shorter, and a lot of treasury processes were designed around a buffer that no longer exists.

The CLS cut-off is the hard edge

The specific constraint most firms underestimate is the submission deadline for payment-versus-payment settlement through CLS, which sits at midnight CET for next-day settlement. PvP matters because it removes the settlement risk of paying out one currency and not receiving the other; it is the safe way to settle FX. The problem under T+1 is that trades executed late in the day — particularly by investors in Asia-Pacific or the Americas dealing in European or US securities — can miss that midnight cut-off. Miss it, and the choice is to settle the FX bilaterally, outside PvP, taking on principal settlement risk, or to fund the position by other means.

When the US moved to T+1 in May 2024 the feared migration out of CLS did not materialise at scale; CLS volumes held up and the share genuinely at risk of missing the cut-off was small. That is reassuring, but it is not a reason to leave it unmanaged. The firms that came through cleanly did so because they identified in advance which of their own flows sat close to the cut-off and re-timed them. The ones that did not are the ones that ended up settling bilaterally on the day, discovering their exposure at the point of failure rather than in a design review. This is a resilience question before it is a treasury one: the tighter cut-off is a fixed edge in your operating day, and your funding process either fits inside it or it does not.

Securities lending: the recall is now the bottleneck

The other half of the funding problem is inventory. If you sell a security that is out on loan, you have to recall it in time to deliver against the sale. Under T+2 the recall notice, the return of the stock and the delivery all fit inside two days with room to spare. Under T+1 the recall has to be issued and actioned on a schedule that leaves almost no tolerance — issue the recall late, or let it sit in a batch overnight, and the stock will not be back before the delivery is due.

This is where the compression bites hardest, because lending recalls have historically been a same-day, business-hours, sometimes manual process. The moment a sale is known, the recall clock starts, and it is now short enough that a recall triggered by an overnight batch or a next-morning review will already be too late. The fix is not more staff on the desk. It is making the recall trigger event-driven off the sale rather than schedule-driven, so the notice goes out as the position is sold rather than when someone next looks at the book. Firms that do not automate this end up borrowing to cover, buying in, or failing — and an unreturned loan that causes a delivery fail is exactly the kind of avoidable exception a shorter cycle punishes.

Where the fails and buy-ins actually come from

Put the two together and the pattern is clear. A settlement fail under T+1 is far more likely to originate in funding — currency that missed the cut-off, stock that came back late — than in the securities instruction itself. That matters because the instruction is where most readiness effort has gone, and it is not where the residual risk sits. The settlement-readiness programme for the middle office gets the trade to the custodian on time; it does not, on its own, guarantee the cash or the stock is there to meet it. Those are separate desks, separate systems and separate cut-offs, and they have to be re-timed deliberately.

The consequences are not abstract. A fail under the CSDR discipline regime carries cash penalties that accrue daily, and persistent fails carry the threat of mandatory buy-in. If you are instrumenting the exception queue for CSDR penalties, tag the funding-driven fails distinctly, because they have a different root cause and a different fix from an instruction mismatch. And if you settle across the EU, UK and Switzerland, remember that a single flow can touch three rulebooks in one coordinated cutover — the FX leg does not respect the boundary between them.

What to re-time, concretely

The work is to map your own funding and recall timelines against the new cycle and find where they no longer fit. For each currency you fund in, ask what time the FX must execute to make the CLS cut-off given your trading day, and whether your current process reliably clears it — for the flows furthest from the settlement currency, it may not. For securities lending, ask whether the recall is triggered by the sale event or by a schedule, and move it to the event. Then decide, in advance and in writing, what happens to the flows that still cannot make the cut-off: pre-funding, bilateral settlement with the risk understood and accepted, or a change to how and when those trades are executed.

None of this is exotic. It is the unglamorous re-timing of two processes that worked fine with a day of slack and will not work without it. The firms that treat T+1 as purely a middle-office deadline will discover the funding gap on 12 October 2027. The ones that treat it as a resilience problem will have found it two years earlier, in a review, where it costs nothing but attention.