How Instant Settlement Affects Treasury and Liquidity

Treasury functions at banks and large corporates have, for most of their history, run on a tempo that is set by the settlement layer below them. Securities settle T+2 (T+1 in the United States since 2024), correspondent banking settles overnight, card schemes settle daily, and the liquidity, reconciliation and funding practices of treasury reflect those windows. Payments move during the day, but their financial impact lands when the scheme settles. Treasury's daily rhythm — forecast, fund, sweep, reconcile — is a rhythm imposed by the plumbing.

Instant settlement changes the tempo. When the underlying rail settles in seconds — domestic instant payment schemes, stablecoin flows, RTGS on DLT, tokenised securities that settle atomically with cash — the treasury function has to operate continuously rather than in daily cycles. The buffers that used to compensate for settlement lag become dead capital; the reconciliation that used to run overnight has to be ambient; the intraday liquidity lines that used to be sized for known cut-offs have to be sized for unpredictable demand. At the same time, instant settlement creates opportunities — precise cash management, dynamic funding, just-in-time hedging — that treasury teams with the right infrastructure can realise.

This article sets out the operational and financial consequences of instant settlement for treasury, and points to the places where engineering investment pays back. The audience is the treasury engineer, treasurer, or architect responsible for the treasury management system at a bank or a large fintech.

The formal difference between T+0 and T+2 is three calendar days. The substantive difference is the decoupling of the payment event from the settlement event. Under T+2 a securities purchase generates a payment obligation that settles two days later; the portfolio is updated on trade day, the cash moves on settlement day. Treasury knows the payment will happen, forecasts it into the daily funding, and has the intervening time to arrange the funding. Under T+0 the payment and the settlement are the same event: cash moves when the trade executes, and the funding has to be in place when the trade happens.

Several regulated infrastructures now operate at T+0 for at least part of their flow. Domestic instant payment schemes (FedNow, TIPS, Faster Payments, UPI, Raast) settle retail and low-value corporate payments in seconds. Tokenised money market funds and stablecoins settle transfers in seconds on public chains. RTGS-on-DLT schemes — which we covered in our RTGS on DLT pillar — settle wholesale flows in seconds with absolute finality. And the US securities market has moved to T+1 for most instruments with T+0 visible on the horizon.

For treasury, the operational consequence is that "end-of-day" stops being a meaningful time boundary for an increasing fraction of flow. The book has to be in balance continuously rather than at a cut-off. The daily sweep that used to move balances to a concentration account has to be continuous or event-driven. The reconciliation that used to run against a batch of transactions has to run against a stream of events.

The first financial impact is on liquidity buffers. Under T+2 a bank holds buffer liquidity to fund payments that are known to be coming but have not yet settled. The buffer is a function of the settlement lag: the longer the lag, the larger the buffer, because more payments are in flight at any given moment. Under T+0 the payments that are in flight become zero-duration — they settle the moment they happen — and the buffer required to cover them shrinks accordingly.

For a large bank the reduction can be material. Industry estimates (which should be treated as indicative rather than precise) put the intraday liquidity held for settlement purposes at a few percent of daily settlement volume. If the settlement volume is in the tens of billions, the liquidity held is in the hundreds of millions to low billions, and the opportunity cost of that liquidity — the spread between the rate it earns in the collateral pool and the rate it could earn if deployed — is the economic case for moving to T+0.

The case is not automatic. Instant settlement introduces its own buffer requirements, because the payments have to be fundable at the moment they happen rather than at the end of the day. A bank that previously funded settlement at the evening cut-off has to fund it continuously, which means intraday credit has to be available on demand. The treasury engineering is to replace the long-duration buffer with a short-duration, dynamic funding capability; the net result is less capital tied up, but only if the dynamic funding exists.

Corporate treasuries see a similar pattern. A corporate that pays suppliers via instant rails can reduce the working-capital buffer it holds for payables, because the moment of payment is controllable to the minute. The buffer reduction is economically significant for corporates with thin margins; for capital-light fintechs and e-commerce companies, it is material to unit economics.

The second impact is on reconciliation. Under batched settlement, reconciliation is a periodic process: match the bank's statement against the internal ledger, investigate breaks, adjust. Under instant settlement, reconciliation can be ambient: each payment event generates a ledger entry that reconciles against the scheme's record in seconds, and breaks surface as they happen rather than being discovered hours later.

The engineering pattern is event-driven reconciliation. The treasury system subscribes to the scheme's event stream (via webhooks or a Kafka topic, depending on the scheme) and processes each event against the internal ledger. An incoming event that matches an expected outgoing payment reconciles cleanly; an event without a matching internal record raises a break; an internal record without an incoming event times out after a bounded wait and raises a break. The backlog of unreconciled items, which under daily batching can run into the thousands, shrinks to a handful of active breaks at any moment.

The ambient reconciliation pattern depends on each payment carrying structured context through the settlement event back to the treasury system. This is where ISO 20022 earns its place — the structured remittance, reference and purpose fields survive the settlement and are available to the reconciliation logic without a separate lookup. Our ISO 20022 and tokenised settlement post covers the data layer in more detail.

For teams building this capability, the main engineering investment is the internal ledger itself. A ledger that was designed for end-of-day reconciliation (write posting, reconcile on statement arrival, finalise) does not naturally support the stream-at-a-time model. The conversion is non-trivial but pays back quickly once the volume of instant-settled flow is meaningful.

The third impact — and the one that most interests treasury innovation teams — is the emergence of intraday credit as an operational capability. Under T+2 a bank arranges overnight funding and expects it to cover the next day's settlement; intraday credit is a specialised product for occasional liquidity needs. Under T+0 intraday credit becomes a structural part of the funding stack, because the moment of maximum funding need can be any second of the day and cannot be predicted with certainty.

Dynamic funding takes several forms. An instant payment rail that allows participants to pre-fund a settlement account at a central bank effectively converts reserve balances into intraday liquidity. A repo market that clears continuously (rather than overnight) allows treasuries to fund positions as they develop. A token-based liquidity facility — for example, a tokenised money-market instrument that can be redeemed for cash in seconds on chain — makes the intraday liquidity profile smoother and reduces the need for dedicated credit lines.

A cross-border corollary is that instant settlement reshapes correspondent banking. A correspondent that settles on a T+2 schedule requires pre-funding of nostro accounts in each corridor, with the nostros sized for the settlement lag. A correspondent that settles instantly — through a stablecoin leg or through an RTGS-on-DLT scheme — requires nostros only for the moment of settlement. The capital released is meaningful; Bridge's own view is covered in the cross-border payment infrastructure pillar.

The final, and most enabling, consequence is the scope for treasury automation. When settlement is continuous and reconciliation is ambient, the treasury's decision loop can be machine-driven rather than human-driven.

Cash concentration can run on events rather than schedules: when a regional account crosses a threshold, sweep to the concentration account automatically; when the concentration account is above target, invest into a liquidity fund; when forecast funding is needed, draw from the fund. Each decision is deterministic and auditable, and the loop runs at whatever cadence the event stream provides. A treasury system that used to run three sweep cycles a day runs many tens.

Hedging can run on positions rather than end-of-day snapshots. As foreign-currency payables or receivables settle, the treasury automatically marks them, nets against existing hedges, and places incremental hedges via an FX platform. The drift between the economic exposure and the hedge exposure is bounded to seconds rather than to a full business day.

Investment allocation can run continuously. Excess cash is automatically allocated to tokenised money-market funds or on-chain liquidity instruments with intraday redemption, earning return on idle balances that would otherwise have sat in non-interest-bearing operational accounts. As demand for cash rises through the day, the allocations redeem progressively.

The common pattern is machine-to-machine integration with the settlement layer. Bridge exposes its settlement events over the three surfaces — REST, webhooks and Kafka — so a treasury system can operate at any cadence appropriate to the flow. The engineering cost is the integration work and the decision logic; the payback is in reduced buffers, reduced reconciliation cost, and better yield on working capital.

Learn about Bridge settlement on /settlement, explore the product surface on /platform, or contact the team to discuss treasury automation for your organisation.