Cut-off times: the hidden clock inside correspondent banking

Every institution in a payment chain runs on a business day, and every business day ends. The moment it ends for new payments is the cut-off time, and for cross-border payments it is the single most underrated source of delay: not a fee, not a technology, just a clock that says tomorrow. A cross-border payment does not face one cut-off. It faces one per institution it touches, each set in a different time zone, and the delays compound. This page is about how that compounding works and how to plan around it.

One payment, many deadlines

Consider a payment moving through a classic correspondent chain: the sender’s bank, an intermediary bank in the currency’s home market, the beneficiary’s bank. Each stage has its own deadline, built backward from the settlement systems it uses: the payment system’s closing time, minus internal processing, minus a compliance buffer. The customer-facing cut-off a treasurer sees is the first deadline of several, and clearing it guarantees nothing about the ones downstream.

The deadlines interact with the payment’s other frictions. A sanctions screening query that takes three hours does not cost three hours; it costs three hours plus whatever windows close in the meantime, which is how a brief hold becomes a full day. A payment that arrives after the intermediary’s deadline queues overnight in a nostro account that had to be funded in advance to receive it.

The geometry of time zones

Direction matters. A payment traveling west gains hours: an afternoon instruction from Singapore reaches a New York correspondent before that bank’s morning, comfortably inside its day. A payment traveling east loses them: a New York afternoon is already evening in London and the middle of the night in Asia, so an eastbound payment released after lunch has often missed every deadline it needed until the next business day. Treasurers who move money in both directions learn the asymmetry quickly: the same corridor can behave like a same-day rail one way and a two-day rail the other.

Friday to Tuesday, step by step

The canonical worst case runs like this. A US company releases a supplier payment to Asia on Friday at 5pm Eastern, after its bank’s cut-off. Processing begins Monday. The payment reaches the intermediary bank Monday New York time, which is already past the Asian beneficiary bank’s Monday. Value arrives Tuesday, local time. Four calendar days have passed. Nothing failed, nobody was slow; every institution processed the payment within its published windows. Add a Monday holiday in any of the three countries and the arrival slips to Wednesday.

How treasurers plan around the clock

Payment operations teams manage cut-offs the way logistics teams manage sailing schedules:

• Release early, and in the corridor’s morning. The cheapest fix is initiating payments in the receiving region’s business morning, which usually means the sender’s prior evening or a scheduled overnight batch.
• Keep a cut-off calendar. Banks publish per-currency cut-offs; sophisticated treasuries consolidate them, per correspondent, with local holidays layered in, and set internal deadlines an hour or more ahead of the bank’s.
• Match the value date to the promise. A supplier promised value Friday needs an instruction that clears every window by Thursday; treasurers work the chain backward.
• Pre-position liquidity. Where deadlines cannot be beaten, balances are held in-country or in-currency so obligations settle from local funds, at the cost of capital trapped in transit.

All of this is skilled, necessary work, and all of it is overhead imposed by the structure of the rails.

What 24/7 rails change

Rails that never close dissolve the deadline at the settlement layer. Instant payment systems run continuously; central banks have been extending RTGS operating hours, and the FSB’s 2025 monitoring reports the share of RTGS systems operating 24/7, and of systems completing an extension, increasing. Shared-ledger rails settle any hour of any day, with finality at 3am on a Sunday identical to noon on a Tuesday.

What remains is the boundary. A payment that settles instantly on an always-on rail still touches institutions that keep business days: the account funding it may sit in a batch system, the currency conversion may wait for market hours, the compliance review may wait for an analyst’s morning. The clock has not disappeared; it has moved to the edges, and it keeps its force wherever the payment’s path re-enters a business-day system.

Where Frame fits

Frame is the settlement layer for global finance: one integration to orchestrate payments at scale across fiat rails, stablecoins, and tokenized deposits, with compliance enforced on every transaction and settlement in seconds instead of days.

Cut-offs are a property of the rail a payment happens to take. Frame is rail-neutral: it routes each payment across whichever rail fits the corridor, the counterparty, and the policy that governs it, which means the routing decision can weigh the clock itself, an always-on rail for the Friday evening payment, the scheduled fiat rail where the deadline is comfortably clear. The compliance checks that turn brief holds into missed windows run inside settlement instead: Frame’s Rules Engine evaluates every transaction against its governing policies as a condition of settling, so a payment that satisfies them does not sit in a queue waiting for a business day, and one that cannot does not leave. For banks, payment providers, exchanges, platforms, and enterprises, that turns cut-off management from a planning discipline into a routing parameter.

See how a rail-neutral settlement layer approaches this in the Frame Blueprint.