The good news: the European Central Bank thinks the world will enjoy cheap, instant and secure cross-border payments within the next decade – what it calls the industry’s “holy grail”.
The bad news: there is a big fat question mark. How do we get there? The news is disappointing if you believe in a purely market-driven approach, but it’s encouraging to those favoring a role for central banks.
That’s because, as the authors note, while digitization has made cross-border communications essentially free, there hasn’t been a commensurate reduction in costs when it comes to payments, despite the advent of fintechs and crypto.
Getting payments into shape
The authors, Ulrich Bindsell (director general at ECB responsible for market infrastructure and payments) and George Pantelopoulos (economist at University of Newcastle in Australia), explain why cross-border payments remain costly and clumsy – and give their view on what solution can do the heavy lifting to change that equation.
First, intermediaries must deal with heavy-handed AML and KYC compliance. Second, while payment providers have made the front-end user experience much more convenient, the back-end operations and payments infrastructure remain impervious to quick fixes.
These include costs associated with currency conversion, network costs, correspondent costs, liquidity and credit costs (especially when forced to work with local banks in foreign jurisdictions), and costs of transparency to ensure against money laundering and other evils. These high costs mean financial institutions don’t want to service low-ticket clients.
Correspondent banking barbell
Payments could become faster and cheaper by streamlining compliance-related processes – but that’s out of this paper’s expertise. But these costs have led to a decline in correspondent banking services – in availability and in quality. This has led to initiative such as better messaging standards (such as the introduction of ISO20022, a data-rich protocol, and improved processes at SWIFT, the global messaging system for correspondent banks).
- Read more:
- Central banks turn to digital to assess new risks
- State Street Digital puts faith in CBDCs
- The BIS agenda: Asia CBDCs and tokenization
Correspondent banking has universal reach, but it will remain a complex, layered set of banking relationships, with a fee and paperwork at every turn. The status quo also remains exposed to the IOUs among participating banks, and the risk of a bank insolvency. Experience shows the industry will always be dominated by a handful of big global players with pricing power.
So what about fintechs? Using agile tech stacks, new entrants have been able to reduce costs in the market segments they serve – think WorldRemit or Wise. Their use of digital infrastructure, remote user identification, and lack of brick-and-mortar costs has enabled them to beat banks in areas such as remittances.
However such fintechs struggle to keep their costs down when they enter new market segments, such as serving corporate treasuries. Most operate as closed-loop networks, restraining their ability to become universal providers, while depending on the infrastructure already built by banks and classic payment processors. They are too siloed to provide a universal path to the payments holy grail, but they have shown what digital tech can make possible.
But who needs service providers at all when you’ve got Bitcoin? Add on service layers like the Lightning Network, and boom: instant, global, 24/7, quasi-free payments, with any FX conversations done separately by each user, and no risk of a bank going down in the middle of a transaction.
There are downsides, however, say the authors. The underlying Proof of Work technology is wasteful and expensive. Bitcoin relies on the entire network to validate blocks and transactions, which may be secure, but is actually more expensive and less scalable than the traditional method of relying on legal systems and compliance.
The low costs of crypto will also disappear as these assets and their (centralized) wallet providers come to be regulated more like financial institutions. The volatility of Bitcoin also makes it unsuited to payments, including at the domestic level.
(The ECB paper did not look at Ethereum and its plan to shift to Proof of Stake, nor at other blockchain-based payment networks such as Ripple. DigFin suggests they should – maybe Bitcoin is already archaic!)
If stability is what the ECB wants, then why not try a stablecoin? When Facebook proposed Libra/Diem, its stablecoin was almost universal, efficient, easy to add on KYC/AML elements, and bound to a basket of major fiat currencies. What’s not to like?
Well, the authors didn’t like giving a BigTech company access to so much sensitive data, potentially handing a private company excessive market power. They are skeptical about a tech company’s ability to manage its reserves balances, or the ability of central banks to surveille them. The ECB is also worried about a popular stablecoin squeezing out sovereign currencies (especially in emerging markets), threatening macro stability, and fragmenting global payment markets into buckets of closed-loop, competing tech ecosystems.
Benchpressing domestic systems
Perhaps then it’s better to return to what we know works. Many countries have introduced faster payment systems that enable instant, real-time settlement for domestic transactions. Is there a way to knit these together? Stitch on a universal FX converter and presto, you’ve got your cross-border holy grail.
There are already steps in this direction. Singapore and Thailand piloted the first cross-border mesh of their domestic payment systems in 2021. This year, five Southeast Asian markets are joining their mobile payments systems via QR codes, for residents traveling among these countries. The Bank of International Settlements is proposing a network called Nexus that would provide the universal FX layer.
The ECB authors like this. It leverages existing domestic infrastructure. The FX part is governed by an international utility, leaving banks to compete on other aspects of service. The design is pretty straightforward. Once a country joins, the service is accessible to all. And central banks aren’t threatened by having a new currency undermine their fiat.
But…something like Nexus isn’t just built by faceless technocrats. It requires national legislatures to support it. The industry has to agree to a body such as BIS or SWIFT to operate it (and given the recent use by the West to sanction Russia and others using SWIFT, not everyone will sign up for this).
Interlinking may be politically feasible in regions such as Southeast Asia or the Nordic countries, but it could struggle to become universal – either because blocs don’t want to play in the same sandbox, or because the costs of servicing smaller, marginal markets will be higher.
Central bank deadlift
This leaves one final path to the payments holy grail: the central banks save the day! (This is an ECB paper after all.) They argue central-bank digital currencies, with a Nexus-like FX service layer on top, may prove the best solution.
CBDCs preserve monetary stability, are straightforward to connect because it eliminates the need for converting money held in commercial-bank accounts, and compliance features should be easy to plug in.
The only catch: there aren’t any CBDCs from major economies. Even China’s eRMB is still at a pilot stage, while no G10 central bank is ready to launch.
The authors argue for central banks and commercial banks working together to launch both CBDCs and linkages among domestic faster payment systems. Both will need the FX utility, so might as well get cracking on that. A bit of competition, as well as a resilient backup, is good. Because central banks will need their CBDCs to work first in a domestic setting, it will be several years before any are ready for international purposes (although China may surprise them), so linking domestic infrastructure can make headway now.
For those interested in the details, including plenty of geekery on how cross-border payments work in these scenarios, access the paper here.