Cross-border pilots between central banks

Sending money to another country today usually passes through a chain of banks. The sender's bank hands the payment to a partner bank, which may hand it to another, until the last bank credits the receiver. Each link keeps its own books and checks the payment in its own way, so the trip can take time and the sender cannot easily follow it. Details such as cut-off times and local holidays add further delay. Central banks and researchers have therefore tested whether a single shared platform could replace part of that chain. The idea is that banks in several countries would settle directly with one another on the same ledger, using digital money issued by their own central banks. Such tests are experiments, and what they show is limited by design.

One well-known test is Project mBridge, run by the Innovation Hub of the Bank for International Settlements together with several central banks. In 2022 it ran a pilot of about six weeks with commercial banks from four jurisdictions. The banks used a platform built on a shared ledger, where payments and currency exchange could settle in one step, with each country's central bank digital money recorded side by side. The reason for this design is that two parties exchange value on the same ledger at the same moment, which removes the waiting between separate books. Participants made real-value transactions for their clients under the pilot's rules, so the test went beyond a laboratory demonstration.

Picture a small exporter named Tomas who sells furniture to a customer abroad. In today's chain, his customer's bank sends a message to a partner bank, which forwards it, and Tomas sees the money only after each link has done its checks. In a pilot like mBridge, the customer's bank and Tomas's bank would both sit on the shared platform. The customer's bank would pass central bank digital money in one currency, the platform would exchange it for the other currency, and Tomas's bank would receive it as a claim on its own central bank. The sequence is shorter, and each party sees the same record at the same moment, so there is less to reconcile afterwards.

The report that followed the pilot said it showed potential for faster and cheaper payments across borders and for lower settlement risk, meaning less chance that one side pays and the other does not. It also listed policy, privacy and governance questions that remained open. Who sets the rules of a platform that several countries share? Which banks may join, and who decides? How is customer data protected when several authorities have a view of the ledger? What happens when national laws disagree? A pilot cannot answer these, because it runs under temporary rules agreed for the experiment, with a small group of willing participants who chose to be there.

A reader who wants to judge such news can ask a few plain questions. Was it a pilot or a production system, and how long did it run? How many banks and countries took part, and were their customers involved? Does the report describe measured results or only expected ones? Which questions does the report itself say remain open, such as access, governance and legal frameworks? Is the platform compared with other routes, since banks, fintech firms and stablecoins all compete to carry cross-border payments? A pilot with open questions is a stage in research, and a headline that treats it as a finished replacement for banks skips those questions entirely and can leave readers with a false picture.

The primary reading here is the Bank for International Settlements report on Project mBridge, which sets out the design, the pilot and the open issues. The same institution has also published a working paper that compares competing routes for cross-border payments, including banks, fintech firms and stablecoins, and that gives a wider view of the choices. Neither document predicts what will happen next, and this page does not either. It is general education, not financial or legal advice, and it does not rank any payment route. Readers who want current details should check the publishing bodies directly, since pilots end, change shape or move to new phases without much notice.

• A pilot tests an idea under temporary rules; it is not a production system
• A shared platform could settle both sides of a payment at once, which shortens the usual chain of banks
• Open questions on governance, access, privacy and law decide how far such systems can spread