Central banks across Asia have spent the better part of a decade running retail central bank digital currency (CBDC) pilots — China's e-CNY, India's e-rupee, and smaller programmes in Hong Kong and Thailand among them. The emphasis is now shifting toward a harder problem: whether digital currencies issued by different monetary authorities can settle payments directly with each other, bypassing the layered network of correspondent banks that currently routes almost all cross-border transactions in the region.
That shift is visible in a wave of wholesale and cross-border experiments coordinated in part through the Bank for International Settlements (BIS) Innovation Hub, which has run or supported multi-CBDC bridge projects involving monetary authorities in China, Hong Kong, Thailand and the United Arab Emirates, among others. Unlike the retail pilots, which let ordinary consumers hold and spend a digital version of the domestic currency, these wholesale trials are built for banks and other financial institutions to exchange central-bank money directly across borders on a shared settlement layer.
From Consumer Pilots to Interbank Plumbing
The distinction between retail and wholesale CBDCs matters more than it might first appear. Retail digital currencies compete, at least partly, with commercial bank deposits and mobile payment apps — a crowded space in most of Asia already, given how deeply platforms like Alipay, WeChat Pay and India's UPI are embedded in daily life. Wholesale CBDCs operate one level up: they are digital claims on the central bank that only banks and licensed institutions can hold, used to settle large payments between each other rather than to buy a coffee.
Several Asian monetary authorities have said openly that their retail pilots produced useful data on consumer adoption and offline payment resilience but ran into a ceiling — domestic payment rails in most of these markets were already fast, cheap and widely used before a CBDC existed. Cross-border settlement is a different story. A trade payment from a manufacturer in Vietnam to a buyer in South Korea can still pass through two or three correspondent banks, each charging a fee and adding a business day or more, with the final settlement amount exposed to exchange-rate movement between the time the payment is sent and the time it clears.
The Correspondent-Banking Problem
Correspondent banking — the system in which banks hold accounts with each other to process international payments — has been the backbone of global finance for decades, and it works. It is also slow and layered by design: a payment typically needs a chain of intermediary banks, each performing its own compliance checks, before funds reach the final recipient. Industry estimates commonly cited by the BIS and national payment associations put average cross-border settlement times at one to five business days, with total fees that can run into several percentage points of the transaction value for smaller transfers.
Multi-CBDC bridge projects aim to compress that chain. Instead of a payment hopping between several correspondent accounts, participating banks hold and exchange digital central-bank currencies on a common technical platform, with foreign-exchange conversion happening atomically alongside settlement rather than as a separate step. In theory, that removes the settlement-timing mismatch that creates FX risk during a cross-border transfer — money and currency conversion happen in the same instant rather than a day or two apart.
Trade finance is the sector most frequently cited as the near-term beneficiary. Letters of credit, invoice financing and other trade instruments currently depend on paper-heavy verification chains that can take a week or more to clear for a single shipment. Faster, programmable settlement — where a payment releases automatically once a shipping document is verified on a shared ledger — is the scenario most pilots have been built around, even though full commercial rollout of that kind of automation remains some years off across most of the region.
A Patchwork of National Approaches
No two Asian CBDC programmes look alike, and that is partly the point of the region's various pilot structures — each is testing a different design question before anyone commits to a single standard. China's e-CNY has the largest user base by far, distributed through state-owned banks and integrated with existing mobile-payment infrastructure, and Beijing has signalled interest in using it for settlement with trading partners along the Belt and Road corridor. India's e-rupee pilot, run by the Reserve Bank of India, has focused more narrowly on wholesale interbank settlement and programmable retail use cases such as targeted subsidy disbursement.
Hong Kong's Monetary Authority has positioned its e-HKD work explicitly around cross-border and wholesale use, reflecting the territory's role as a financial hub rather than a mass retail payments market. Thailand's central bank has similarly leaned toward wholesale and cross-border testing through its long-running Inthanon project and its successor initiatives, often in partnership with Hong Kong. Japan and South Korea have moved more cautiously on retail CBDCs — both have run limited pilots without committing to nationwide issuance — while contributing technical expertise to multilateral bridge experiments rather than pushing ahead unilaterally.
What It Means for Commercial Banks
Commercial banks are watching this shift carefully, and not entirely comfortably. A wholesale CBDC that lets central banks settle directly with each other, or with large corporates, narrows the role correspondent banks have traditionally played as fee-earning intermediaries in cross-border payments. Industry groups representing regional banks have generally welcomed the efficiency case while pushing central banks to preserve a role for commercial institutions in the new settlement chain — as liquidity providers and compliance gatekeepers, if not as the payment routers they are today.
Disintermediation risk is more acute on the retail side, where a widely adopted digital currency held directly with the central bank could, in theory, draw deposits away from commercial banks during periods of financial stress — a scenario central banks in Asia have generally tried to design around by capping individual holding limits or routing distribution through existing banks rather than issuing CBDC directly to the public. Wholesale CBDCs sidestep that particular risk since only licensed institutions can hold them, which is one reason the region's central banks have found it easier to build political and industry consensus around the wholesale track first.
Interoperability Remains the Real Obstacle
The technical challenge that keeps coming up across these projects is not building a CBDC — most central banks in the region can now do that — but making systems built by different countries, on different technology stacks, talk to each other reliably. Some pilots use distributed-ledger technology, others use more conventional centralised databases with shared messaging standards, and reconciling the two approaches into a single cross-border bridge has proven slower than the initial pilot announcements suggested it would be.
Regulatory fragmentation compounds the technical one. Capital-flow controls, anti-money-laundering requirements and data-localisation rules differ sharply between, say, China and Singapore, and a settlement bridge has to satisfy all of them simultaneously rather than the more permissive standard any single jurisdiction might apply on its own. That is one reason most of the multi-CBDC bridge projects so far have stayed in pilot or proof-of-concept phase for multiple years rather than converting into always-on production infrastructure — the hard part was never proving the technology works, but agreeing on who bears liability and oversight when it does.
Privacy and Capital-Control Concerns Persist
Skepticism about CBDCs in Asia tends to split along two lines. Privacy advocates and some retail users have raised concerns, particularly around China's e-CNY, that a digital currency issued and tracked directly by the central bank gives the state a more granular view of individual transactions than cash or even existing digital payment apps provide. Central banks involved in these pilots have generally responded by building in tiered anonymity for small transactions, though the underlying architecture still allows for transaction-level visibility that cash never offered.
A separate concern, more relevant to the wholesale and cross-border track, is capital-flow management. Several economies in the region maintain some form of capital controls, and instant, programmable cross-border settlement could make it technically easier to move large sums quickly — a feature central banks want for legitimate trade finance and a risk regulators want to guard against for currency stability. Reconciling those two goals within the same settlement bridge has been one of the slower-moving parts of the multilateral projects, and it is unlikely to be resolved through pilot programmes alone.
Outlook
None of the current cross-border CBDC bridges has moved from pilot to full production settlement infrastructure, and central bank officials involved in these projects have generally avoided giving firm timelines for when that transition might happen. What has changed is the direction of travel: the early framing of CBDCs as a retail payments innovation has given way to a wholesale and cross-border settlement agenda that more directly targets the frictions in correspondent banking and trade finance. Whether that translates into a functioning alternative to the current system, or simply a faster, better-documented layer sitting alongside it, is likely to become clearer only as more of these bridge projects move past the pilot stage over the coming years.