Canadian blockchain project urges banks to proceed with caution
Despite the real-world use cases being found around blockchain and distributed ledger technology, banks are being warned to be cautious about committing to any such projects.
Canada's Project Jasper, a research initiative focusing on blockchain and what role it might have in central bank transaction settlement, is advising banks to take a wide view of the pros and cons of distributed ledger technology and not just assume it can automatically improve systems that currently work fine.
Payments Canada, the Bank of Canada, the R3 blockchain consortium and major banks have worked on Project Jasper for more than a year, testing various aspects of the technology that was initially developed to manage and track bitcoin transactions through a ledger visible to all and maintained by a distributed system of miners.
"Looking at distributed ledgers in which only authorized entities can make payments or perform functions on the system, we found at a very high level that we didn't really see that there was much promise in terms of net benefits" to a bank, said Scott Hendry, senior special director overseeing financial technology research at Bank of Canada.
Even in just looking at the costs of operating a wholesale payments system, it is not clearly evident that the distributed ledger would be a benefit over a centralized system for inter-bank payments that are already in place, Hendry said last week during the annual Payments Symposium at the Chicago Federal Reserve Bank.
Those involved with the proof-of-concept Jasper Project and its various phases have concluded that a standalone blockchain wholesale payment system is unlikely to replicate the benefits of a centralized wholesale payment system, but it does have some potential in interacting with broader aspects of the financial market infrastructure.
One such example would be adding cash on the same ledger as payments involving exchange-traded assets, in order to settle the "cash leg" of each transaction.
"We are modernizing the payments system in Canada, but it is not being done because it is inefficient," Hendry said of the country's faster payments initiative. "It is being done because it is an old platform and we want to add functionality."
Still, the key lesson learned from Project Jasper is that banks need to look at it more broadly than as just a core payments system to determine where the payoff might be for distributed ledger technology, Hendry added.
Project Jasper will continue to research the potential for blockchain in a cross-border payments setting.
"I applaud the Payments Canada folks for following through with Project Jasper," said Peter Tapling, chief commercial officer at SpringLabs, a company building blockchain-based protocols for financial services and other industries.
"As an industry, we are never going to know how some of this stuff works unless somebody follows it through all the way to the end," Tapling said.
A plus for blockchain is that it doesn't present a major learning curve for banks in that it is not a technology entirely foreign to the industry.
"We clearly understand the cryptographic transforms that are being used in blockchain, so that's good that we don't have to relearn that," Tapling said. "But the new things are the value chain and the consensus model, and we do have to wrap our heads around that."
Tapling said it is good for a bank's technology team to have researched new innovation and has a clear focus on what problems it is trying to solve and has already formulated a technology it believes might work.
"In speaking to many" of the banks' executives, "it seems that on average it can cost $1 million just to get a project started because they need a project manager, a legal person and a security person" along with IT, Tapling said. "It is hard work to dig into the research and see if a new technology works for them."
Hendry acknowledged the process is vital, but that even Project Jasper is only testing what could be, or might be.
"We go into these phases with several hypotheses that we want to test," Hendry said. "The unfortunate part is these are proofs of concept, rather than real production systems."
In that regard, the researchers are "only getting an indication of what the answer would be as opposed to hard data," he added. "The answers are qualitative, rather than quantitative."