Carolyn A. Wilkins is the senior deputy governor of the Bank of Canada. Gerry Gaetz is the president of Payments Canada.
A year ago, Payments Canada, the Bank of Canada, major Canadian banks and technology consortium R3 began an ambitious project – code-named Jasper. Its goal was to build and test an experimental wholesale interbank payment system, using distributed ledger technology (DLT) – the backbone of cryptocurrencies such as Bitcoin.
The wholesale payment system is a crucial, yet not well-known, part of the financial system's foundation. Operated by Payments Canada and overseen by the Bank of Canada, it is used to transfer large and time-critical funds. On average, more than $175-billion a day is cleared through the current Large Value Transfer System in Canada.
Meanwhile, DLT – a decentralized electronic database shared between multiple parties – is now being considered and tested across a range of financial applications. Wholesale payments systems appear to be natural candidates for DLT applications, given their relative simplicity. And thanks to recent enhancements, DLT is now better designed for this critical infrastructure.
So the natural question is: Could DLT underpin an entire wholesale payment system?
The answer is maybe one day, but there remain many hurdles to overcome.
Let's start with the "maybe one day."
Our experiment, done in two phases, demonstrated that it's indeed possible to settle wholesale payments on a distributed ledger. Our work also found that such a system could meet some, but not all, of the core international principles for financial market infrastructure. In the second phase of the project, we even successfully incorporated an innovative liquidity savings mechanism – a queuing process that greatly increases efficiency by allowing institutions to net some of their payments. And while cost savings wouldn't likely come from the core system, distributed ledger platforms could potentially enable savings by lowering back-office reconciliation costs.
Now for the hurdles. Several important gaps and fundamental challenges remain.
One is the need for privacy around transactions in wholesale payments systems. This is incompatible with some versions of decentralized digital ledgers – which operate under an assumption that everything is publicly observable at a certain level.
While we were able to address this privacy constraint in our second phase, the fix made the system susceptible to the risk of a single point of failure – something present in current systems but that was supposed to disappear with distributed ledgers.
Another hurdle was scalability, which is still an issue in some versions of DLT. While other versions can achieve greater transaction rates by moving away from a fully decentralized framework, this can reduce resiliency and lessen some of the expected potential cost savings.
The bottom line is that a stand-alone DLT wholesale system is unlikely to match the efficiency and net benefits of a centralized system. In fact, at its heart, there exists a fundamental inconsistency or tension between a centralized wholesale interbank payment system, as we have now, and the decentralization inherent in DLT.
At the end of the day, interbank systems must be safe, secure, efficient and resilient, and they must meet all international standards. DLT-based platforms are just not there yet.
Perhaps the biggest lesson from Project Jasper is just how valuable public-private sector collaboration can be to gaining understanding and moving things forward. We encountered and overcame a number of complex technical challenges, and we were able to work effectively together, yielding important insights into each others' perspectives along the way.
There are a host of further avenues that may be worth exploring, and these sorts of experiments offer a collaborative way forward to do just that. Our present view is that the biggest net benefits, if any, would likely lie in the interaction of a DLT-based wholesale payments system with broader financial market infrastructure.
Therefore, one area for future research and experimentation would be to think about broadening out collateral pledging beyond just cash. Another would be to explore the integration between Project Jasper and other types of DLT-based financial market infrastructures, domestically or internationally. This could yield efficiency increases from the ability to settle multiple assets (such as bonds, or money market instruments) on the same ledger.
While these represent opportunities further down the road, Canada's payments system needs to be modernized in the short term. Though it won't involve distributed ledgers, it will involve a lot of innovation and collaboration.
To that end, Payments Canada and private-sector financial institutions, with the full support of the Bank of Canada, are already under way on a multiyear initiative to modernize Canada's payments system. Our shared vision is of a system that's fast, flexible, promotes innovation and strengthens Canada's competitive position. And above all, we need a system that is safe and secure, so that Canadians can count on a financial system with solid foundations.