For BRICS, Connecting Payment Systems Is Only Half the Challenge

Prior to the September 12–13 BRICS Summit in New Delhi, India was urging members to link their payment systems, with connections between central bank digital currencies (CBDCs) reportedly on the agenda. This ambition is not new, with several discussions happening on the need to hedge against US dollar hegemony and the potential development of a BRICS currency.

The more consequential question, however, is not whether money can move faster, but rather but who supplies the liquidity, who recognizes the settlement and who absorbs the loss when something goes wrong. After all, trust is the fundamental lifeline of cross-border payments and the wider financial system. Rather than requiring just a notification, an exporter needs usable funds, predictable conversion costs and certainty about when payment becomes final. Importantly, connecting national systems does not automatically deliver those things.

Three Projects, Three Different Choices

Multilateral and bilateral efforts have been under way for some time to explore what the right model for a BRICS cross-border system might look like (see Table 1). Project Nexus — in which India, along with the Bank for International Settlements (BIS), is participating — is designed to connect existing domestic instant-payment systems through standardized links. It is not yet operational, but crucially, this approach does not require countries to introduce a common currency or launch CBDCs first.

Project mBridge — a BIS-developed initiative — takes a different route. It suggests a need for a shared ledger for cross-border wholesale transactions in multiple CBDCs, with participants that include China, Hong Kong, Thailand and the United Arab Emirates. That said, it is speculated that the BIS supposedly exited the project due to geopolitical considerations over the heavy involvement of China and its aims to internationalize the digital yuan.

Lastly, Project Agorá is another initiative of the BIS, in conjunction with 40 other financial institutions, which combines tokenized commercial-bank deposits with tokenized central bank reserves. It conducted real-value testing in July 2026 but remains in an experimental stage. Its significance is architectural, given that its focus on modernizing international payments does not mean removing commercial banks, which would change the landscape of the common two-tier banking system.

Each of these solutions provides a different take on the same problem, but each has its own limitations. For the BRICS implementations to be successful, it must be able to distinguish everyday retail connectivity from wholesale settlement reform. Making every case dependent on CBDC adoption would unnecessarily narrow its options.

The Liquidity Problem Survives the Technology

Interlinking different systems also creates failure paths, as one system, for example, may debit funds before another rejects the payment. In that case, the Nexus system requires domestic operators to guarantee settlement and handle reversals, including when certain participants default. Compatibility means agreeing how to recover from failure, not merely how to exchange messages.

While a common messaging standard, such as ISO 20022, can help systems exchange information, it cannot make international laws agree. Atomic, payment-versus-payment settlement can ensure that both currency transfers occur or that neither does, addressing the risk of paying without receiving. Even so, technical completion still needs legally enforceable finality across jurisdictions, and the availability of atomic settlement does not eliminate exchange-rate risk or create liquidity.

Consider an exporter receiving a currency it cannot readily spend, invest or convert. A faster transfer has changed the location of the problem, but it has not solved it. Persistent trade imbalances still require someone to hold the surplus currency or finance the deficit. Central bank swaps may provide temporary liquidity, but they cannot substitute indefinitely for demand for the currency and its assets. Even elements such as operating hours matter, as Nexus’s own documentation mentions foreign-exchange providers charging less competitive rates at weekends, when conventional currency markets are closed and providers must cover the risk of prices moving. Thus, an instant payment is not necessarily a cheap currency-conversion substitute.

As such, the geopolitical implications of this competition are significant. A recent National Bureau of Economic Research working paper models currency competition around financial-market liquidity and network effects, and suggests a limit to what payment technology alone can achieve. The strategic objective for the BRICS members is not simply to reduce the use of the US dollar, but to avoid replacing one concentrated monetary dependency with another. For example, if Chinese institutions become the dominant foreign-exchange dealers, market makers or emergency liquidity providers within a BRICS payments architecture, the renminbi could gradually become the system’s de facto anchor even without being formally designated as a common currency. That would increase China’s influence over conversion pricing, liquidity conditions and potentially affect the resilience of participating economies during periods of financial stress, while also creating new asymmetries between China and smaller BRICS members.

Conversely, if local-currency transactions continue to be priced, hedged or funded through dollar markets, the new interface may reduce the visible use of the dollar without materially reducing dependence on US financial-market liquidity or exposure to dollar funding conditions. In other words, currency competition affects not only transaction costs, but also monetary autonomy, bargaining power and the distribution of financial risk within the bloc. For BRICS nations, genuine diversification requires multiple sources of liquidity and market-making capacity so that payments integration does not simply shift dependence from Washington to Beijing, or obscure the dollar’s continuing centrality.

Control Moves but Does Not Disappear

The same principle applies to sanctions. US Treasury guidance explicitly states that certain Russia-related transactions can expose foreign banks to sanctions even when conducted in non-dollar currencies. Changing the payment route does not remove that exposure. The risk is that a permissive common rulebook could deter globally active banks, while a restrictive one could exclude the participants most eager for alternatives.

BRICS can’t avoid answering another integral question: Who controls the infrastructure? Who admits or suspends a bank, approves software changes, accesses transaction data or orders an emergency shutdown? Who compensates customers after a cyberattack? Shared infrastructure without agreed authority can turn an operational incident into a diplomatic confrontation. Member states should negotiate rights of oversight, appeal and orderly exit as carefully as technical access. They should also demand a design that lets them switch technology and liquidity providers without disrupting payments; otherwise, dependence may simply migrate from correspondent banks to platform operators.

The distinction between retail and wholesale access also matters for financial stability. A International Monetary Fund working paper warns that access to a safer foreign CBDC can intensify domestic deposit flight. That is not a prediction about every wholesale bridge, but rather a reason to specify precisely who may hold foreign digital money, in what quantities and under which safeguards.

Following the BRICS summit this year, Delhi should seek to establish a payments rulebook and not simply advocate for greater connectivity between national payment systems. Retail links and wholesale settlement experiments should proceed separately, each with a defined purpose and participating institutions capable of meeting their obligations. The rulebook must be technology-agnostic and should not prescribe whether payments must ultimately run through CBDCs, linked instant-payment systems, tokenized deposits or shared ledgers. Instead, it could establish common requirements for interoperability across these architectures, including messaging standards, identification and authentication, foreign-exchange interfaces, settlement processes, data exchange and mechanisms for handling technical failure.

Following that, every corridor should also specify conversion pricing, committed liquidity, settlement finality, data access, fraud redress and responsibility for losses before scaling. Tests should include the failure of one payment technology to communicate or settle reliably with another. Lastly, success should be measured by the total cost of funds arriving usable and final, alongside interoperability, reliability and recovery from failure, rather than by headline transaction speed.

In the long run, the ultimate prize is not a payment system free of politics, but one whose dependencies are visible, negotiable and sufficiently diversified. Connecting payment systems is the visible half of the challenge. The harder half is everything that makes those connections trustworthy: committed liquidity, enforceable finality, accountable governance and credible plans for when things go wrong. Without a rulebook that answers those questions, faster payments simply move the same problems across borders.