Why it matters
Delivery versus payment is the oldest and most consequential piece of settlement risk management in securities markets. Without it, one side of a trade must part with value first, exposing the full principal amount to the other side's failure. The Committee on Payment and Settlement Systems formalised the concept in its September 1992 report, prompted by the concern that disturbances in securities settlement could spread to payment systems and to financial markets generally. It is now an obligation rather than an aspiration: the CPMI-IOSCO Principles for financial market infrastructures require that a system settling two linked obligations eliminate principal risk by conditioning the final settlement of one on the final settlement of the other.
Tokenization has revived the topic because it changes how the linkage is achieved, not whether it is required. Most claims made for blockchain settlement are claims about implementing delivery-versus-payment with fewer intermediaries.
How it works
The 1992 CPSS report identified three structural approaches, and the taxonomy still organises the field. In the first, securities and cash both move trade by trade on a gross basis, so each transaction is independently final. In the second, securities move gross while cash obligations are netted and settled at the end of a cycle. In the third, both are netted, and final transfers occur once per cycle.
The distinctions determine what a participant is exposed to between trade and settlement. Gross-gross settlement removes principal risk continuously but demands that both assets be in place at the moment of each trade. Netted models economise on liquidity and on the number of transfers, at the cost of an intraday exposure and a dependence on the netting cycle completing.
On tokenized platforms the linkage is enforced by code and by the location of the cash leg. Where a platform can hold tokenized central bank money, settlement occurs in the safest available asset. Where it cannot, the cash leg is a commercial bank deposit token, a payment system trigger, or a stablecoin, each of which changes the credit quality of the settlement asset even when atomicity is identical.
Economic mechanism
Delivery versus payment is best read as a choice about where to place a cost. Eliminating principal risk requires either collateral and margin, or simultaneity. Simultaneity is not free: it requires the seller to have the asset unencumbered and the buyer to have funds available at the same instant, which raises intraday liquidity demand and reduces the scope for netting.
The BIS Quarterly Review of March 2020 set the argument out directly: tokenized systems gravitate toward gross settlement, which "generally requires more liquidity", and participants may therefore not want shorter settlement cycles even where the technology permits them. The corollary is that the case for tokenized delivery versus payment is strongest where intraday liquidity is cheap relative to the credit exposure being removed: short-dated repo, collateral transfers, and cross-border transactions with long settlement chains.
A second mechanism is the value of certainty itself. When settlement is conditional and final, positions can be re-hypothecated, financed or margined with less buffer. Collateral velocity, not settlement speed, is where the measurable saving usually appears.
Participants
Central securities depositories and central counterparties remain the institutional home of delivery versus payment. Central banks supply the cash leg and the legal finality; the Swiss National Bank's Project Helvetia tests settlement of tokenized assets on a delivery-versus-payment basis using a wholesale central bank digital currency on the SIX Digital Exchange platform, runs until at least June 2028, and admits only institutions holding a sight deposit account at the SNB. Custodians, triparty agents, platform operators and clearing members hold the operational obligations. The Eurosystem's Pontes track, announced on 1 July 2025 with a pilot targeted for end-Q3 2026, exists to allow DLT platforms to settle in central bank money.
Examples
Project Helvetia pursues two designs in parallel: an integrated approach, in which wholesale central bank digital currency is issued onto the DLT platform, and a synchronised approach, in which the platform is linked to the existing real-time gross settlement system. That is the clearest live illustration of the trade-off between putting the cash leg on the asset ledger and triggering it in a separate system.
At scale, tokenized repo shows what delivery versus payment looks like in production. Broadridge reported that its Distributed Ledger Repo platform processed nearly $9 trillion of notional in December 2025, with average daily volume of $384bn, up 490% year on year. That is monthly turnover of largely overnight transactions, not outstanding balances.
Risks and limitations
The most common error is to treat atomicity as sufficient. If the cash leg is a private claim of uncertain quality, delivery versus payment removes principal risk on the exchange while leaving the buyer holding an asset whose value depends on an issuer's solvency.
Legal finality does not follow automatically from ledger finality; it depends on designation under national settlement finality regimes, which is why platform-level legal analysis features so heavily in central bank pilots. Fragmentation is a second limitation: a capability that works only inside one platform narrows the pool of eligible counterparties and can raise the total cost of settlement. Moving from netted to gross settlement also transfers risk into the intraday liquidity domain, where it is less well measured.
Key metrics
The measures that matter are settlement fails as a share of value settled, intraday peak liquidity usage, the share of settlement value where the cash leg is central bank money rather than a private claim, and the value settled on a linked delivery-versus-payment basis on tokenized platforms, the $384bn average daily volume Broadridge reported for December 2025.
Related research
The CPSS report of September 1992 defines the models; the CPMI-IOSCO Principles state the standing obligation. The BIS Quarterly Review article of March 2020 is the reference for the liquidity trade-off, and Project Helvetia and the ECB's Pontes and Appia programmes are the most advanced central bank efforts to supply the cash leg.