Interbank and international payments with blockchain?
What happens when we send a payment across the border? From Switzerland to the USA?
Multiple steps happen within and between banks in order to execute this payment, which are costly and time consuming. Let’s imagine that Mr. Müller has a Swiss franc (CHF) account with bank X and wishes to send 10$ to M. Smith, that has a USD account with bank Y located in the USA. What happens for this payment transfer to go through?
The CHF account Mr. Müller is debited by 10,05CHF (current exchange rate), and 10$ are implicitly bought at bank X. The CHF position of bank X increases while the USD position decreases by 10$.
Bank X has a USD account at a commercial bank (bank Z) in the USA, which is called “X’s correspondent bank”, to whom it credits the 10$.
Bank Z debits 10$ from X’s account at their bank and transfers the money to bank Y, where M. Smith will finally see her 10$.
Step 3 might not be as straight forward. Banks from the same country may have a so-called Nostro account, which allow them to transfer money from each other accounts directly. However, banks also have accounts at a central bank (Federal Reserve in the US) which are mirrored in a local account at the bank, where they credit and debit the central bank. Central banks are used as settlement authorities for the payments in the currency for which they are responsible.
Therefore, in case bank Z does not have a Nostro account with bank Y, the Federal Reserve (FED) needs to be involved to settle those 10$.
10$ are such a low amount of money that additional parties need to take part, such as clearing houses, so that low value payments can be grouped, and the central bank does not have to be involved in every interbank payment.
For Mr. Müller to send 10$ to M. Smith via bank transfer, 5 entities need to get involved, which makes the whole procedure time consuming and costly. We made a comprehensive graphic where we outline every step of an international bank transfer (Figure 2).
How can distributed ledger technology make interbank and international payments more efficient?
As for any database, the entities that write state in a database are called writers. In blockchain, this refers to the participant that is involved in the consensus protocol and helps growing the network: a writer accumulates transactions within a block and appends this block to the blockchain. On the other hand, the entities that do not extend the blockchain but participate in the transaction process are called readers.
A system of interbank payments has multiple parties, the banks, that can act as writers: they create transactions. Within the same country, or within countries having the same currency (European Union, for example), the central bank functions as a trusted party, and could also act as a certificate authority giving out licenses to banks to be able to participate in the system. Therefore, all writers of the system are known and a permissioned blockchain could be used.
In order for different currency systems to function together, and for international payments to work as smoothly as interbank payments may work with blockchain, the different currency blockchains should be designed in a way that interactions between chains were possible. The ultimate goal would be to provide atomic cross currency payments (either all transactions go through, or none of them).
The main challenge for adopting distributed ledger technology (DLT) for interbank and international payments is that the margins earned by banks would decrease. That is the reason why this solution for upgrading the corresponding banking system faces some resistance for adoption. It is true that banks could also save costs on human power, as all transactions would be automated, but they would certainly earn less fees.
Furthermore, setting a common worldwide interbank and interbank payment system powered by DLT would require very high switching costs to shift from the current corresponding banking system to a new blockchain-based one. For that reason, these blockchain solutions need to offer clear advantages such as greater speed, cheaper transactions, more transparency and maintain security in cross-border payments, and at the same time, ensure that banks can still be competitive while taking away their margins. The objective would be that as many banks as possible sign up for DLT initiatives for international payments, so that we can reach mass adoption and develop a worldwide payment system with all the benefits that blockchain can offer.
Where are we at today?
Some central banks such as the Monetary Authority of Singapore (MAS) and the Bank of Canada are working on solutions to use DLT for interbank payments. In Singapore, banks would be able to deposit some amount of money with the MAS and would receive the same amount on the distributed ledger. Then, banks can use the ledger to immediately transfer money between the banks without having to use a Nostro Bank account with the central bank. This does not allow to execute cross-border transactions, but it is the first step for DLT-powered interbank payments.
A similar use case exists for the global treasury teams of large corporations: it can be complex for multinational treasurers to have a real time view of liquidity, as business demands that enterprises maintain local accounts for various subsidiaries with multiple banks around the world. The company Adjoint offers a solution to this problem with their product Smart Treasury. Thanks to blockchain technology, treasurers can have round-the-clock visibility of consolidated bank balances, liquidity, investments and positions across multiple banks, divisions, countries and currencies. It also allows for zero-cost transactions within and outside the organization.
Moreover, JPMorgan has built a blockchain-powered cross-border payment product, the Interbank Information Network (IIN), and as of the 25th of September of 2018, 75 international banks have signed up to test it. As JP Morgan announced in a press release, “IIN minimizes friction in the global payments process, enabling payments to reach beneficiaries faster and with fewer steps. Using blockchain technology, IIN reduces the time correspondent banks currently spend responding to compliance and other data-related inquiries that delay payments. IIN is powered by Quorum, a permissioned-variant of the Ethereum blockchain, developed by J.P. Morgan.”
IIN competes with legacy platforms such as SWIFT and new startups like Ripple. The latter aims to provide a global settlement network based on a distributed ledger. It only partially replaces the correspondent banking system as banks can continue to use correspondent banks to process payments. Similar to the traditional correspondent banking system, a payment may require multiple steps if no trust relationship exists between the two banks that are parties in the transaction. However, the payment is atomic (either all of the intermediate payments go through or none of them). In the traditional system, if something goes wrong for an intermediate payment, previous payments have to be reversed and manual intervention may be required. Ripple provides its own currency, XRP, which can be used as intermediate currency for transactions.
It is unclear how long it will take until we see international payment transactions powered by distributed ledger technologies. On one hand, some central banks are working on DLT solutions to alleviate some of the worst pains that business and private persons currently must endure when sending money overseas. Also, other initiatives such as the International Information Network are uniting international banks to work on new blockchain-based payment systems. However, adoption remains the main challenge as in order to achieve real-time settlement using blockchain tokens, both commercial and central banks need to get involved. The technology is not quite ready yet neither, and a lot of research and development needs to be done for having different currency blockchains being able to interact with each other, and to achieve atomic cross currency payments.
Do you need blockchain — Wüst & Gervais (2017)