The Last Look…

Payments and FX have always had a rather strange relationship: on the one hand, cross-border money movements is the very reason for the FX and its market structure idiosyncrasies. On the other, an institutional trading desk rarely troubles itself with the frankly boring, mechanical task of transferring funds from Customer A to B. 

This perception of dullness hides one of my favourite stats for this year, namely that the payments industry remains the most valuable part of financial services, generating $2.5 trillion in revenue from $2.0 quadrillion in value flows, supported by 3.6 trillion transactions worldwide.

In an era of margin compression and rising costs, the “most valuable part of financial services” might be an area of interest, especially as the whole cross-border element of it turns on currencies. Since the advent of stablecoins, payments has become a lot more interesting from an FX perspective, but there is still a major disconnect between the two spaces, it seems, even when it comes to high-level policy makers.

After six years of international effort to fix cross-border payments, policymakers appear to have discovered something the wholesale FX market has known all along: moving money across borders means that at some point, somebody has to change the currency.

This point was made unusually explicitly by Martin Moloney, deputy secretary general of the Financial Stability Board, in a recent speech looking ahead to what comes after the G20’s Cross-Border Payments Roadmap.

The roadmap was launched in 2020 with the commendable ambition of making cross-border payments faster, cheaper, more transparent and ultimately more inclusive. This is a noble aim as currently the vast majority of costs are shouldered by the smallest and least-well off communities and nations. Despite the stated goals, the efforts have not yielded much success so far, in fact, some metrics show a deterioration in outcomes. Moloney acknowledges that there remains considerable distance to travel and identifies what he calls “exogenous dependencies” as one reason why.

Like FX, for example. Cross-border payments, he says, “rely heavily on foreign exchange markets”, yet those markets sit outside the scope of the roadmap. 

This points towards a fundamental problem with attempts to redesign international payments: the payment, the FX transaction and the settlement of that FX transaction cannot easily be considered in isolation.

The FSB itself identifies FX access restrictions, payment-versus-payment coverage and operating hours among the outstanding frictions affecting cross-border payments, hardly peripheral issues for the FX industry. They go directly to liquidity, credit, settlement risk and the ability to deliver currencies when and where they are required.

And technology in itself does not make the FX leg disappear either.

Stablecoins are the obvious example. They are frequently presented as a way of bypassing slow correspondent banking chains and moving value across borders almost instantaneously. But Moloney offers a useful reality check.

The FSB estimates total cross-border payments at around $200 trillion in 2024. By comparison, some estimates put stablecoin cross-border payments at less than 0.2% of the total in 2025. Moloney’s conclusion is that stablecoins remain a very small part of the market and may initially prove more useful as components of hybrid systems, integrated with bank money and interoperable FX and settlement infrastructure, than as standalone global payment rails. 

This is where the discussion becomes much more interesting for FX.

A dollar stablecoin might allow dollars to move around the clock, but it does not automatically provide someone in Brazil, Japan or Indonesia with reais, yen or rupiah. Somewhere in the process there remains a conversion between currencies, along with questions about where liquidity resides, who makes the price, how credit is managed and how the two sides of that transaction settle.

Tokenisation may change the mechanics considerably. The BIS’s Project Agorá has already demonstrated atomic settlement of wholesale cross-border transactions using tokenised commercial bank deposits and central bank reserves across currencies and jurisdictions. The project is now moving towards testing involving real-value transactions. 

The BIS argues that tokenisation could collapse sequential processes (payment instructions, compliance checks and settlement) into a single, coordinated workflow, as well as reducing pre-funding requirements and improving intraday liquidity management. That would be significant. 

But again, this does not eliminate FX, it simply changes where and how currencies are traded and converted. And this is why, in a world of interoperable tokenised deposits, stablecoins and conventional bank money, the interesting question becomes who will provide the conversion mechanism connecting these different forms of money.

There are already signs that policymakers are thinking in these terms. The UK and US recently committed to greater regulatory cooperation around digital assets and stablecoins, explicitly linking that work to cross-border connectivity and reducing market fragmentation. 

BRICS countries are discussing interoperability between domestic fast-payment systems and potentially CBDCs, while Brazil is examining international connections for Pix. All of which suggests that the next phase of the cross-border payments project may have to bring FX much closer to its centre.

New rails can make money programmable, tokenised and available 24/7 and they can shorten chains of intermediaries and potentially allow simultaneous settlement. But international commerce still requires one form of money to be exchanged for another. As the universe of instruments that need trading expands, the world’s most liquid market might find a plethora of new opportunities. 

As is so often the case, the technology might change, but the need for an FX market doesn’t.