The international payments network Swift has a target painted on its back, and a number of upstarts are aiming for it. But a new technology it is experimenting with could be the shield Swift has been looking for.
But lately Swift, which once considered Ripple and other tech company rivals an existential threat, has struck a more confident tone. Swift says the progress it is making with so-called global payments innovation, or GPI — technology that lets banks see where their payments are at all times, and that comes with rules around response and confirmation times — will neutralize the competitive threat posed by Ripple and others. It says its bankers are happy with the speed and insight GPI gives them.
Contrary to Ripple CEO Brad Garlinghouse’s previous claims, SWIFT’s global payments innovation (GPI) could soon shake up the competition. GPI is reportedly now in use at 450 banks, which process 80 percent of international payments. Its daily amount of settled payments exceeds $300 billion.
Is GPI the answer to go against Ripple?
What makes SWIFT’s GPI attractive to bankers is the pace and transparency level of its transactions. “Our banks are quite happy with the API-based technology we have in GPI, it produces a faster payment experience,” said Wim Raymaekers, head of the global banking market at SWIFT.
While GPI is not capable of speeding up the payments, it does so indirectly by ensuring they are confirmed by banks within the same day.
Because of this SWIFT believes that blockchain isn’t the solution for a faster payment experience.
According to Raymaekers, speed and transparency are two of Ripple’s main concerns, which blockchain does not seem to solve. Back in 2018, SWIFT ran a distributed ledger experiment to see if switching to a hyperledger would increase transparency of cross-border payments. Soon, they ran into a different issue. “Banks have indicated they do not want to communicate how much money they have on an account to everybody else,” said Raymaekers. “We had to keep the conversations private.”
Moreover, Swift argues, distributed-ledger technology does not fix the delays in international payments that typically occur for two reasons. One, legal and regulatory requirements make the use of a shared ledger impossible among banks that lack mutual know-your-customer relationships. Second, banks are unwilling to let their rivals see how much money they have.
By the end of the experiment, SWIFT concluded that the only way to upkeep an acceptable level of transparency would require deploying bilateral blockchains between all parties, but doing so would be too complex to manage in the long run. “Our conclusion is that it’s quite challenging for banks to adopt a DLT-based system today,” said Raymaekers.
Swift’s blockchain experiment
Asked if Swift would ever use a distributed ledger for the movement of payment messages within its network, Raymaekers said, “Never say never, but we haven’t yet seen how DLT would improve the service we are currently providing to customers.”
Last year, Swift and 40 member banks ran a blockchain proof of concept for so-called nostro reconciliation.
Swift banks open nostro (from the Latin for “our”) accounts with each other to facilitate cross-border payments. The proof of concept tested the idea of using a hyperledger to see the positions of those accounts in real time. If an account is empty, the bank that receives the payment instruction would not be able to pay the beneficiary.
“What banks need to do is to improve the speed by which they communicate the status of the account to you,” Raymaekers said. “They realize that they have to change their back-office processes in order to provide more frequent updates.”
The other problem with using a blockchain was an excess of transparency. Each participant could see every update on the ledger.
“Banks have indicated they do not want to communicate how much money they have on an account to everybody else,” Raymaekers said. “We had to keep the conversations private.”
Swift ended up deploying bilateral blockchains between all the participants because the banks did not want others to know how much money was in their nostro accounts. Running several hundred bilateral blockchains became very complex.