Alternatives to correspondent banking for cross-border settlement

Most cross-border payments still travel through correspondent banking: a chain of banks holding accounts for one another, each adding cost, delay, and an operational cut-off. The search for alternatives is no longer speculative. The BIS itself has written that existing correspondent banking processes “have struggled to adapt” to regulatory requirements, “posing questions on the future of the correspondent banking model”, and the network is contracting, with active correspondent relationships down about 22% between 2011 and 2019 even as payment volumes grew.

Meanwhile the official improvement programme is behind schedule. The G20’s targets call for retail cross-border payments to cost no more than 1% on average by end-2027, with no corridor above 3%, and for 75% of wholesale payments to arrive within an hour. In October 2025 the Financial Stability Board concluded that satisfactory improvement at the global level is unlikely on that timetable, and the World Bank still measures the average cost of sending remittances at 6.36%.

So what are the alternatives? They fall into six groups. This page maps all of them, with what is live and what is not, and then looks at the question the map itself raises.

1. The incumbent system, upgraded

The first alternative to the old correspondent chain is a better correspondent chain. SWIFT gpi tracks payments end to end across the existing network: by SWIFT’s own figures, nearly 60% of gpi payments reach the end beneficiary within 30 minutes, almost 100% within 24 hours, and member banks send over $300 billion a day through it. Visa B2B Connect takes a different route to the same goal, a multilateral bank-to-bank network that Visa markets explicitly as “an alternative cross-border solution” to correspondent processes: one connection, no intermediary fees, irrevocable payments delivered in full value.

These fix visibility and, partly, speed. They do not change the underlying model of pre-funded accounts and chained intermediaries.

2. Interlinked instant payment systems

Domestic instant payment systems work well; the idea of Project Nexus is to connect them to each other, so a cross-border payment reaches the recipient within 60 seconds in most cases. The central banks of India, Malaysia, the Philippines, Singapore, and Thailand agreed in 2024 to take Nexus toward live implementation and set up a managing entity, Nexus Global Payments, in Singapore, with the BIS in a technical advisory role only. This is the most direct public-sector answer for retail and SME payments in the corridors it covers.

3. Bank-owned settlement networks

A group of consortium networks now settles interbank payments on shared ledgers, live, without correspondent chains. Partior, founded by DBS, J.P. Morgan, and Temasek, went live with commercial flows of USD, EUR, and SGD in 2023 and settles in commercial bank money with atomic, 24/7 payment-versus-payment capability; Deutsche Bank completed its first euro cross-border payment on the network in September 2025. Fnality, owned by more than twenty major financial institutions, went further up the settlement-asset hierarchy: its sterling payment system, live since December 2023, settles in balances backed one-for-one by funds held in an omnibus account at the Bank of England. We cover both networks, and the Regulated Liability Network work that became the UK’s tokenised sterling deposit pilot, in detail in the interbank settlement network landscape.

4. Central bank projects

Two BIS-convened projects frame the official-sector frontier. Project mBridge built a common platform on which central banks in China, Hong Kong, Thailand, the UAE, and Saudi Arabia settle in their own digital currencies; it reached minimum viable product in mid-2024, at which point the BIS handed it to the partner central banks, its General Manager noting it was “not mature enough to start operating”. Project Agorá, convened with the Institute of International Finance, is broader: eight central banks and over 40 financial institutions testing tokenised central bank reserves and commercial bank deposits on one platform. In May 2026 it published prototype results demonstrating atomic multi-currency settlement and moved toward real-value testing, while the BIS is explicit that Agorá “is not about building a finished product”.

5. Fintech networks with direct local connections

The fintech answer bypasses the correspondent chain with direct memberships in domestic payment systems. Wise is the clearest proof that this works at scale: $243.5 billion in cross-border volume in its 2026 fiscal year, up 31%, at an average cross-border take rate of 0.52%, with 75% of its fourth-quarter payments completing in under 20 seconds. Ripple Payments offers payouts across more than 60 markets on a network it says has processed over $100 billion, settling in fiat or in regulated stablecoins including its own RLUSD, issued under a New York trust charter. These networks are real alternatives for the flows they serve; they are also closed loops, value moves fast inside the network and hands off to slower rails at the edge.

6. Stablecoin settlement rails

The newest group settles cross-border value in regulated stablecoins, cutting the intermediary chain to a single transfer on a shared ledger. The Circle Payments Network, live since May 2025, coordinates payments between vetted financial institutions that settle directly with each other in USDC and EURC; Circle counts more than $28 trillion in cumulative USDC settlement volume since 2018. Paxos issues regulated tokens for other institutions, PayPal USD among them, under OCC-supervised trust structures. Stripe closed its acquisition of Bridge, a stablecoin infrastructure platform, in February 2025 in a deal reported at $1.1 billion. The full landscape, including bank-issued tokens and the consortium efforts challenging the issuer model, is mapped in stablecoin settlement infrastructure for regulated institutions.

The question the map raises

Lay the six groups side by side and a pattern appears: every alternative is a network, and every network is partial. Each covers certain corridors, certain currencies, certain membership. Each settles a different asset, commercial bank money on Partior, central bank money on Fnality, digital currencies on mBridge, USDC on CPN, and none of them interoperates with the others. An institution that commits to one has traded the correspondent chain’s problems for a new one: betting on which network wins, while its liquidity fragments across venues that do not connect.

And in every case, compliance remains someone else’s problem. The networks move value faster; the checking of the value’s conditions, sanctions, limits, permissions, jurisdiction rules, still happens in systems bolted on before or after settlement.

Where Frame fits

Frame is the settlement layer for global finance: one integration to orchestrate payments at scale across fiat rails, stablecoins, and tokenised deposits, with compliance enforced on every transaction and settlement in minutes instead of days.

Frame’s answer to the map above is not to be another region on it. Frame is rail-neutral: a payment enters through one integration, and Frame routes it across whichever rail fits the corridor, the counterparty, and the policy that governs it, a fiat network, a regulated stablecoin, or a tokenised deposit. Compliance is enforced inside settlement, not around it: Frame’s Rules Engine evaluates every transaction against its governing policies, and a transfer that cannot satisfy them does not settle. Every settled transaction produces verifiable evidence that its conditions were met, without exposing the underlying business data.

For banks, payment providers, exchanges, platforms, and enterprises, that turns the strategic question this page began with, which alternative do we pick, into an operational one: what policy should govern each payment, and which rail serves it best today. The map will keep changing. A settlement layer that spans it does not have to.