What is just-in-time liquidity?

Just-in-time liquidity is a funding model in which cash arrives where a payment settles at the moment it is needed, in the amount needed, rather than being parked in advance against a forecast peak. It is the treasury equivalent of just-in-time inventory: buffers shrink because visibility and settlement are fast and reliable enough to fund on demand instead of on fear.

Is just-in-time liquidity actually new?

The principle is proven inside existing market infrastructure. CLS reduces the cash its settlement members need for FX settlement by roughly 99% through multilateral netting and In/Out swaps, and major RTGS systems such as TARGET2 and CHAPS run liquidity-saving mechanisms that queue and offset payments so banks need less intraday liquidity. What is new is extending that discipline to cross-border corridors that still rely on pre-funded correspondent accounts.

What do treasurers gain from just-in-time funding?

Smaller idle balances. Capital that sat in nostro accounts or local subsidiaries as a buffer against slow, uncertain settlement can be released once settlement is fast and observable. The gain shows up as reduced working capital needs, less reliance on intraday credit and overdrafts, and fewer currencies held simply because moving them took days.

What does just-in-time liquidity require to work?

Three things. Real-time visibility into balances and incoming payments, because you cannot fund on demand if you learn positions from yesterday's statement. Settlement that is fast and final at any hour, because a buffer is only removable when its replacement arrives in minutes. And disciplined cash forecasting with agreed fallbacks, such as intraday credit lines, for the days the forecast is wrong.

Treasury

Just-in-time liquidity

Just-in-time liquidity replaces peak-sized pre-funding with funding on demand. How CLS and RTGS systems proved the model, and what modern settlement rails change.

Just-in-time liquidity is funding that arrives when the payment does. Instead of parking cash in advance against the largest day the forecast can imagine, a treasury funds each settlement obligation at the moment it falls due, in the amount actually owed. Manufacturing learned this lesson decades ago: inventory is a cost, and the better your logistics, the less of it you need. Cash buffers are inventory. The quality of the settlement rails determines how much of it a treasurer is forced to hold.

Today, most cross-border flows still run on the opposite model. The correspondent system settles on pre-funded accounts, so banks and corporates size nostro balances to peaks, not averages. The cost of that model, capital idle across currencies and time zones, is covered in the real cost of pre-funding; this page is about the alternative and what makes it possible.

The system already proved the principle

Just-in-time funding is not a fintech hypothesis. The most conservative infrastructure in finance runs on it.

CLS. FX settlement’s answer to Herstatt risk is also a masterclass in liquidity compression. CLS settles both legs of a trade payment-versus-payment, then multilaterally nets each member’s obligations, and then compresses the remainder further with In/Out swaps, which offset payment obligations by an average of about 75%. The combined effect, per CLS, is that members fund less than 1% of gross settlement value in cash, a roughly 99% reduction in the balances settlement would otherwise demand.

RTGS liquidity-saving mechanisms. Real-time gross settlement removes settlement risk but is hungry for intraday liquidity, so the major systems added machinery to economize it. TARGET2 runs offsetting algorithms continuously against its payment queues; CHAPS, BOJ-NET, and SIC operate equivalent mechanisms. The Bank of England’s overview of liquidity saving in RTGS describes the design goal plainly: settle the same payments with less liquidity by matching and offsetting what would otherwise queue, and the ECB’s analysis of TARGET2 shows how reservations, limits, and algorithms shape how much liquidity the system actually consumes.

Inside the perimeter of a single system, in other words, the industry long ago decided that parked cash is a design failure. The buffers survive mainly in the seams between systems, where cross-border payments live.

What makes just-in-time possible across borders

Three capabilities have to hold at once.

Visibility in real time. Funding on demand requires knowing positions on demand: balances, incoming payments, and obligations, across banks and currencies, now rather than at yesterday’s statement. This is the least glamorous requirement and the one that kills most buffer-reduction plans.

Settlement that is fast, final, and always on. A buffer exists because its replacement is slow. When a transfer achieves finality in minutes at any hour, including the weekend the correspondent chain sleeps through, the case for holding peak-sized balances collapses. Atomic settlement strengthens the case further: when linked obligations settle all-or-nothing, no party needs a buffer against the half-completed state.

Netting before funding. The cheapest liquidity is the payment you never make. Multilateral netting collapses gross obligations into net positions first, so just-in-time funding applies to a fraction of the original flow, which is exactly the CLS sequence: net first, fund the sliver.

Honest requirements and limits

Just-in-time liquidity is an operating discipline, and it has real preconditions. Cash forecasting has to be good enough to trust, with intraday credit or committed lines as the fallback for the days it is not. Counterparties and banking partners must support the cadence; a corridor whose off-ramp keeps banking hours re-imposes the buffer at the boundary. Cut-over is gradual in practice: treasurers shrink buffers corridor by corridor as evidence accumulates, rather than switching models on a Monday. And some cushion always remains, because operational risk does not disappear just because settlement got faster.

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.

For treasury, Frame works as an overlay on existing bank rails, with no stablecoin required: netting and pre-rail settlement reduce the obligations that need funding at all, and settlement in seconds replaces the peak-sized buffer with funding sized to the day’s net position. Because Frame’s Rules Engine evaluates every transaction against its governing policies inside settlement, and every settled transaction produces verifiable evidence that its conditions were met, the auditability that treasurers worry about losing with faster settlement is produced by the settlement itself.

See how a rail-neutral settlement layer works: the Frame Blueprint.

Common questions

What is just-in-time liquidity?: Just-in-time liquidity is a funding model in which cash arrives where a payment settles at the moment it is needed, in the amount needed, rather than being parked in advance against a forecast peak. It is the treasury equivalent of just-in-time inventory: buffers shrink because visibility and settlement are fast and reliable enough to fund on demand instead of on fear.
Is just-in-time liquidity actually new?: The principle is proven inside existing market infrastructure. CLS reduces the cash its settlement members need for FX settlement by roughly 99% through multilateral netting and In/Out swaps, and major RTGS systems such as TARGET2 and CHAPS run liquidity-saving mechanisms that queue and offset payments so banks need less intraday liquidity. What is new is extending that discipline to cross-border corridors that still rely on pre-funded correspondent accounts.
What do treasurers gain from just-in-time funding?: Smaller idle balances. Capital that sat in nostro accounts or local subsidiaries as a buffer against slow, uncertain settlement can be released once settlement is fast and observable. The gain shows up as reduced working capital needs, less reliance on intraday credit and overdrafts, and fewer currencies held simply because moving them took days.
What does just-in-time liquidity require to work?: Three things. Real-time visibility into balances and incoming payments, because you cannot fund on demand if you learn positions from yesterday's statement. Settlement that is fast and final at any hour, because a buffer is only removable when its replacement arrives in minutes. And disciplined cash forecasting with agreed fallbacks, such as intraday credit lines, for the days the forecast is wrong.

Sources

Last reviewed 2026-07-16