The mechanics of cross-border settlement have long required capital to sit idle. Before a payment can move between jurisdictions, correspondent accounts must already be funded. This is the nostro trap: money parked across currencies and banking relationships, earning nothing, deployed nowhere, held purely to ensure that when a settlement instruction arrives, the funds are already there to clear it.
The Bank for International Settlements estimates that more than $27 trillion sits idle in correspondent banking accounts globally, capital that cannot be lent, invested, or redeployed because it must remain available for settlement. For individual institutions, the figures are substantial. A large global bank may hold $10-25 billion in nostro balances. Multi-entity corporate groups face the same structural constraint at smaller scale: fragmented liquidity across currencies and banks, operationally slow to reallocate, and politically charged when subsidiaries compete for the same capital pool.
Pontes changes the proof point. On September 21, the Eurosystem launched its platform linking distributed ledger technology platforms to TARGET Services, enabling wholesale transactions in tokenized assets to settle in central bank money. This is the safest settlement asset available, reserves held at the central bank itself, carrying no counterparty credit risk. The platform went live with Deutsche Bank, Santander, Société Générale, and the European Investment Bank among its initial participants, alongside DLT operators including Clearstream and Axiology.
The significance is not that tokenization exists, but that central bank money now moves across it. Previously, tokenized trades settled in commercial bank money or stablecoins, carrying credit risk that large institutions were reluctant to accept at scale. ECB Executive Board member Piero Cipollone framed the launch directly: central bank money "will give an important advantage to help it scale." The ECB reinforced this commitment by announcing it will invest a portion of its own funds in tokenized securities settled through Pontes, targeting euro-denominated debt issued by governments and supranational institutions.
The platform builds on extensive technical groundwork. Between May and November 2024, the Eurosystem conducted exploratory work with 64 participants, processing over €1.5 billion across more than 50 trials and experiments. That programme tested three interoperability solutions for settling DLT-based transactions in central bank money, and the results directly informed Pontes. On July 1, 2025, the ECB Governing Council approved a dual-track strategy: Pontes as the near-term solution, with a longer-term initiative called Appia focused on building a fully integrated DLT-based capital markets ecosystem.
Pontes currently operates during European business hours, 8:00 to 16:00 Central European Time, with round-the-clock settlement and multi-currency capabilities planned for 2028. The initial scope is deliberately narrow: wholesale transactions between eligible credit institutions, market infrastructure providers, and central banks. Retail payments remain the domain of the separate digital euro project, which enters its pilot phase in 2027.
For corporate treasury, the operational relevance is indirect but consequential. Pontes is infrastructure for banks, not corporates. But the infrastructure your banks use determines the constraints you inherit. If your correspondent banking partners begin settling on programmable rails that eliminate pre-funding requirements, the capital you maintain in nostro accounts starts to look less like a necessary operating cost and more like a subsidy for legacy architecture.
The timing sharpens this tension. Pontes launches as corporate treasury teams finalize 2027 planning cycles. The question embedded in those budgets is whether trapped liquidity remains a structural given or becomes a discretionary choice. At current short-term rates, idle capital carries real cost. Balances held for settlement certainty are balances not deployed for yield, not funding working capital needs, not available for opportunistic deployment.
The counter-argument is that Pontes is early, limited in scope, and won't reshape corporate treasury operations overnight. This is true. The platform is a pilot. Operating hours are restricted. The participant list is selective. Migration paths from traditional correspondent banking to DLT-based settlement remain undefined for most institutions. And Pontes addresses the euro-denominated leg of settlement, multi-currency corporate treasury operations still face fragmented liquidity across non-EUR jurisdictions.
But the strategic implication is harder to dismiss. When the ECB itself begins investing in tokenized securities and settling those transactions through programmable central bank money, the institutional legitimacy of this infrastructure shifts. The question for treasury teams is no longer whether programmable settlement can work, but when it will become available through your banking partners, and what you will do with the capital currently locked in pre-funding when it does.
Corporate treasurers have spent years treating nostro pre-funding as an infrastructure cost outside their control. Pontes suggests that characterization has an expiration date.
References
[3] ECB, "Exploratory work on new technologies for wholesale central bank money settlement,"
