RTGS vs Deferred Net Settlement: How Banks Settle Payments

The choice between settling every payment one-by-one or batching them up shapes the safety and cost of the entire financial system. This is the distinction between real-time gross settlement and deferred net settlement — two designs with opposite trade-offs between settlement risk and liquidity efficiency. This guide explains how each works, why both exist, and where each is used.

What problem do settlement systems solve?

When banks make payments to each other on behalf of customers, they accumulate obligations that must eventually be settled — actual funds moved to discharge what each owes. The design of how and when this settlement happens determines two crucial things: how much settlement risk exists (the danger that a bank fails to pay what it owes, leaving others short) and how much liquidity banks need to hold to make payments. The two main designs — gross and net settlement — represent fundamentally different answers to balancing these, and the choice has profound implications for systemic safety.

Settlement design is a technical but pivotal part of the infrastructure covered in our banking hub.

How does real-time gross settlement (RTGS) work?

In an RTGS system, each payment is settled individually (gross) and immediately (real-time), one at a time, by moving central-bank money from the paying bank’s account to the receiving bank’s account. The moment a payment settles, it is final and irrevocable, and the receiving bank can rely on the funds. Because settlement happens in central-bank money with no netting and no delay, there is essentially no settlement risk — no period where one bank is exposed to another’s potential failure. This safety is why RTGS systems are used for large-value, time-critical, and systemically important payments.

How does deferred net settlement work?

In a deferred net settlement system, payments are not settled one by one. Instead, the system collects many payments over a period, nets them — calculating each bank’s single net position (total owed minus total due) — and settles only those net amounts at scheduled times. Because banks send and receive many offsetting payments, netting dramatically reduces the actual funds that must move: a bank might owe and be owed large gross amounts but have only a small net balance to settle. This makes net settlement highly liquidity-efficient, which is why it suits high-volume, lower-value retail payments where settling each individually would be impractical.

What is the trade-off between the two designs?

The two designs sit at opposite ends of a fundamental trade-off. RTGS maximises safety — instant, final settlement in central-bank money means no settlement risk — but demands that banks hold substantial liquidity to settle each payment as it arises, which is costly. Net settlement minimises liquidity needs through netting but introduces settlement risk: between netting cycles, banks are exposed to the possibility that a counterparty fails before settling its net obligation, potentially unravelling the whole batch. Neither is universally better; the right choice depends on whether the priority is eliminating risk (favouring RTGS) or conserving liquidity (favouring net settlement).

How do systems manage the risks of net settlement?

Because deferred net settlement carries the danger that one bank’s failure mid-cycle could leave others short, these systems build in protections. They may require participants to post collateral or contribute to a guarantee fund that covers a defaulting member’s obligation, ensuring settlement completes even if a bank fails. They set caps on how much exposure can build between cycles, and shorten settlement cycles to limit the window of risk. These safeguards aim to capture net settlement’s liquidity efficiency while containing its settlement risk, so that a single failure cannot cascade into a systemic collapse of the netting batch.

How are modern systems blending the two approaches?

The line between the designs is blurring. Modern RTGS systems incorporate liquidity-saving mechanisms — offsetting queued payments against each other before settling, capturing some of netting’s efficiency while keeping gross-settlement safety. Instant-payment systems often settle individual retail payments in real time but use deferred net settlement between the participating banks behind the scenes, combining instant finality for the customer with liquidity efficiency for the banks. The trend is toward hybrid designs that deliver fast, final payments to users while managing liquidity and risk efficiently underneath — a sophistication that reflects how central payment infrastructure has become to a well-functioning financial system.

What is settlement finality and why does it matter?

Settlement finality is the point at which a payment becomes irrevocable and unconditional — the receiving bank can absolutely rely on the funds, and the payment cannot be reversed or unwound even if a participant subsequently fails. Finality is crucial for systemic safety: without it, a chain of payments could be undone if one party failed, spreading losses unpredictably. In RTGS, finality is achieved with each individual payment as it settles in central-bank money, giving immediate certainty. In net settlement, finality comes when the netted obligations settle at the end of the cycle. Legal frameworks underpin settlement finality, protecting completed settlements from being reversed by insolvency proceedings. This legal certainty is what allows the financial system to treat settled payments as truly done, enabling the confident chaining of transactions on which modern finance depends.

How do liquidity-saving mechanisms work in RTGS?

Pure RTGS is safe but liquidity-hungry, since each payment needs funds available the instant it settles. To ease this without sacrificing safety, modern RTGS systems add liquidity-saving mechanisms. Rather than settling every payment immediately in strict order, the system can hold payments briefly in a queue and look for opportunities to offset them — if Bank A owes Bank B and Bank B owes Bank A, the system can settle both together, needing only the net difference in liquidity. It can also optimise the order and timing of queued payments to maximise such offsetting. These mechanisms capture some of net settlement’s efficiency while retaining gross settlement’s finality and safety, letting banks make their payments with less liquidity tied up. This blending is why the strict distinction between gross and net settlement has softened in practice, with sophisticated systems delivering both safety and efficiency.

Who participates in settlement systems?

Direct participation in core settlement systems, particularly RTGS, is typically limited to banks and certain financial institutions that hold accounts at the central bank, because settlement happens in central-bank money across those accounts. Other institutions — smaller banks, fintechs, payment firms — often access these systems indirectly, settling through a direct participant that acts on their behalf. This tiered structure has implications for competition and access: indirect participants depend on direct ones, which can limit their independence and add cost. There is a trend, in some jurisdictions, toward broadening direct access to settlement systems to non-bank payment providers, reflecting how the payments landscape has expanded beyond traditional banks. Widening access can increase competition and resilience but raises questions about risk, since settlement-system participants must meet high standards to ensure the system’s safety, balancing openness against the need to protect critical infrastructure.

How does settlement design affect systemic risk?

The way a financial system settles payments has profound implications for systemic risk — the danger that problems at one institution cascade across the system. Net settlement, by leaving obligations unsettled between cycles, creates interdependencies: if one participant fails before settling, others expecting its payment could be left short, potentially triggering further failures. This is why net settlement systems build in protections like collateral and guarantee funds. RTGS largely removes this channel of systemic risk by settling each payment finally and immediately, which is precisely why systemically critical large-value payments use it. The broader lesson is that settlement infrastructure is not a neutral technical detail but a key determinant of how resilient the financial system is to the failure of any single participant. Designing settlement to contain rather than transmit shocks is a central concern of the central banks and regulators who oversee this critical infrastructure.

How do central banks use settlement systems for monetary policy?

The settlement systems that move money between banks are also instruments of monetary policy and financial stability. Because banks settle in central-bank money held in accounts at the central bank, the central bank can influence the amount and cost of that money, affecting interest rates and liquidity across the banking system. The large-value settlement system is where the central bank’s monetary operations ultimately take effect, and it can provide or withdraw liquidity to support its policy goals and to ensure payments settle smoothly. In times of stress, the central bank can inject liquidity through these systems to keep payments flowing and prevent a liquidity squeeze from becoming a crisis. This dual role — as critical payment infrastructure and as the channel for monetary policy and stability operations — underscores why settlement systems are so central to the financial system and why the central bank’s position at their heart is so important.

What happens when a settlement system participant fails?

The failure of a participant in a settlement system is precisely the scenario these systems are designed to withstand, and the consequences depend heavily on the settlement design. In an RTGS system, because each payment settled finally and immediately, a participant’s failure does not unwind completed payments — others are not left short on settled transactions, which is a key safety advantage. In a deferred net settlement system, a failure before the netting cycle completes is more dangerous, since others may have been counting on the failed participant’s net payment; this is why such systems hold collateral and guarantee funds to complete settlement even if a member defaults. Across both, legal settlement finality protects completed settlements from being reversed by the failed participant’s insolvency. The handling of participant failure is a core test of any settlement system’s design, and the protections built in — finality, collateral, guarantee arrangements — exist specifically to ensure one failure cannot cascade into systemic collapse.

Frequently Asked Questions