Beyond Babel: A New Blueprint for Global Instant Payments – Part 1
Fragmentation has become a very real challenge for the industry as instant payment schemes are being implemented on a country by country basis, and each country is doing it differently. Like the Tower of Babel, the vision of global faster payments is imperiled by mutual incomprehension.
National not international
There are several reasons why schemes are developing on a national rather than a global basis. The obvious one is that despite globalisation, the vast majority of low value payments are made in-country, not cross-border. Deutsche Bank report that even in the Eurozone, with its common currency and payment standards, less than 3% of payments are made cross-border. Germany has the highest share of cross-border payments with 3.1%, whereas cross-border payments account for a mere 0.6% in Italy.
A second reason is the need to scale the network quickly. A successful instant payments scheme must be able to reach most of the accounts in its target market, meaning that most banks must participate from day one. Mobilising banks to co-operate and co-ordinate nationally is difficult enough, and achieving the same thing across national boundaries and multiple currencies is close to impossible.
Is interoperability mission impossible?
Many national instant payment clearing and settlement schemes (CSMs) exist, and many more are in development. The technical standards vary from scheme to scheme, so even if the problems of settlement and multicurrency were resolved (each a huge problem in its own right), the national schemes could not simply “plug in” to each other – there is no interoperability.
The ISO20022 Real Time Payments Group (RTPG) has been working for over a year and has only just published specifications for the most basic message pair needed for instant payments (credit transfer request and response). In order to build a system that can be operated in the real world, many more messages are needed such as administration and control messages to control settlement cycles, participant availability, crypto key management, and a whole host of other vital functions.
We are still years away from having a defined messaging standard, by which time many more non-interoperable national schemes will have launched.
Techno-genies, the value chain and realism
Rather than lamenting the lack of international interoperability, we should consider whether it is even desirable. Imagine that some payments techno-genie granted our wish that all national schemes were magically migrated to technically interoperable systems. What next?
Well, the systems would have to start connecting to each other, and each connection would have to be tested. Routing rules and tables would need to be created and maintained, so each participant knew which banks are reachable via which CSM. The connections would not be purely technical – each pair of schemes would need to put settlement arrangements in place, and operating rules and processes would need to be agreed.
For an instant payment, time is of the essence, and each additional link in the processing chain adds to the time taken to process the payment. It also increases the complexity and hence the likelihood of failures, and makes it more difficult to track down problems that do occur. Hence each CSM should have a direct bilateral connection with every other for settlement tracking/resolution.
Assuming there will soon be 50 instant payment CSMs worldwide each one would need to operate 49 bilateral connections, giving a total of 1,225 separate connections. Every change in payment message specifications would require all of those links to be re-tested, incurring massive costs and a major disincentive to enhancement and innovation. It was precisely to avoid the problems associated with such an explosion of inter-bank connections that national “hub and spoke” CSMs were introduced.
The idea of a global patchwork of bilaterally connected CSMs is not workable and it’s not going to happen. So how can the vision of international instant payments be realised?
To be continued…..