Project Agora signals the future of cross-border payments, exposes banks’ weak spot

By Puja Sharma

As central and commercial banks begin testing Project Agora, the global financial industry is taking another step towards rethinking how money moves across borders. Led by the Bank for International Settlements in collaboration with central banks and private-sector participants, Project Agora is designed to explore how tokenisation and shared ledger technology could modernise wholesale and cross-border payments, a part of the financial system that remains stubbornly slow, costly and opaque.

At its core, Agora examines whether tokenised forms of money, including commercial bank money and central bank-issued money,  can be settled on a unified ledger, reducing reliance on multiple intermediaries and enabling atomic, near real-time settlement across jurisdictions. The initiative aligns closely with the G20 roadmap to enhance cross-border payments and builds on earlier experiments such as Project mBridge and Project Dunbar, marking a shift from narrow CBDC pilots towards a broader, more pragmatic tokenisation framework that actively involves commercial banks.

Yet as these experiments move from concept to user testing, a critical tension is emerging between innovation at the ledger level and the reality of bank payment infrastructure. While tokenisation promises continuous, always-on settlement, many banks — particularly in the US — continue to rely on batch-based processing models built for a very different era of payments.

According to Steve Cook, Co-Founder and Strategic Advisor at Form3, this structural mismatch risks limiting the real-world impact of initiatives like Agora. “Projects like Agora show that tokenisation is fully moving into real-world applications, including cross-border payments. The direction is clear, but the true risk lies in the underlying infrastructure.”

Cook argued that without a move away from legacy, batch-based architectures, banks will struggle to support tokenised assets, including stablecoins, at scale. “It’s impossible to leverage the benefits of tokenisation for cross-border payments while relying on legacy architecture that simply can’t handle them. This is where cloud-native platforms come in. Designed for continuous processing, monitoring and reconciliation, they enable banks and fintechs to deliver fast, always-on payments.”

The distinction between batch and continuous processing is fundamental. Traditional systems process payments at set intervals, introducing delays, reconciliation breaks and limited visibility — constraints that become increasingly problematic in cross-border environments. Tokenised payments, by contrast, require real-time validation, continuous reconciliation and instant settlement across time zones.

“Layering on a new digital instrument on top old tech can create the appearance of progress but will only complicate the already existing problems,” Cook added. This concern echoes a broader industry view that tokenisation cannot compensate for outdated foundations. Tokenisation is not a workaround for legacy systems. Without fixing the foundations, the industry risks running into the same old bottlenecks, delivering innovation that looks impressive on paper but stalls in practice.”

The delivery gap is also highlighted by Pratiksha Pathak, Partner and Head of Payments at RedCompass Labs, who points to uneven progress despite strong regulatory momentum. “Momentum is building around the modernisation of wholesale cross-border payments, with initiatives like Agorá now moving into user testing. Unified ledgers, tokenisation and settlement in central bank money have the potential to materially improve speed, transparency and liquidity efficiency.”

However, she notes that persistent structural issues continue to undermine progress. “On the ground, however, delivery remains uneven. Despite strong G20 and CPMI commitments, cross-border payments are still too slow, too costly and too opaque, and interoperability remains a major constraint.”

Recent warnings from the Financial Stability Board that full global alignment by 2027 is unlikely further underline the scale of the challenge.

Looking ahead, Pathak argued that banks face a narrow window to translate experimentation into production. “2026 is pivotal. Banks must move beyond patchwork fixes and invest in scalable, interoperable infrastructure. ISO 20022, instant payments and real-time cross-border rails require continuous change, not one-off programmes. Those that act now will be better placed to turn innovation into real outcomes.”

For banks watching Project Agora unfold, the message is increasingly clear. Tokenisation may define the future of cross-border payments, but without deep investment in modern, continuously processing infrastructure, that future risks remaining just out of reach — impressive in concept, but constrained in execution.