After a recent column, I got a lot of feedback about the lack of understanding of blockchain, bitcoin and distributed-ledger technology in banking.
So, I reached out to some smart people in the industry and came up with the following list of truths about the adoption of this nascent technology in the industry so far.
After all, misinformation and misunderstandings could be a little dangerous.
“You have to get to understand it or you will make fundamental mistakes,” said Chris Skinner, chair of the Financial Services Club. “You might end up going with Betamax rather than the VHS version.”
Some blockchain projects are little more than bragging rights.
“Blockchain has almost taken on a life of its own in the C suite where it’s become a bit like a CEO vanity project,” said Martha Bennett, principal analyst at Forrester. “What are we doing on blockchain? Can’t we be in the headlines on blockchain?”
Sometimes the chief information officer will have concluded, through testing and research and development, that the technology is too immature, at the same time that the CEO of the same company is demanding a blockchain project.
“There’s a little bit of a cart before the horse thing around that,” she said.
“People are still learning at this point in time about how to use the distributed ledgers and/or blockchain and whether it’s something that’s going to be appropriate for the bank – appropriate for what?” Skinner said. “There’s far too much discussion around, ‘We must be in this space,’ rather than saying, ‘Why must we be in this space?’ ”
The technology is not fully baked.
There is little substance in the early offerings, said Jay Wack, president and CEO of security software company Tecsec. “Crypto and security are better left to folks who understand it. It is not that it cannot be done, it just should be done by those who understand the difference between good sound technology and the latest fad.”
The open-source blockchain products (such as Hyperledger and Ethereum) have missing pieces, Wack said.
“For example, who has access to the ledger?” he said. “That’s done by the server with a user name and password. That’s not a mathematically secured process, it’s a procedure. I’m concerned that people aren’t getting what they think they’re getting. It’s not that it’s wrong but it’s not as good as it could be.”
Many of the blockchains out there can’t handle confidentiality, privacy, performance or scalability, he said.
Blockchain and distributed ledger technology are not the same thing.
Skinner said he is – or at least was – guilty of this one. He used the terms interchangeably on his blog for some time.
“Quite a few experts who have more technical knowledge than I have called me out on it,” he said.
You can have a distributed ledger on a blockchain, but you don’t have to have a blockchain to have a distributed ledger. A distributed ledger also can have, but doesn’t have to have, a digital currency, a digital signature and consensus mechanisms. R3’s Corda distributed ledger, for instance, does not have digital currency or a blockchain.
“It is semantics, but it’s quite important because there’s a delineation between how we develop these technologies and there’s different classes,” Skinner said. “Distributed ledger is a different technical development from the blockchain.”
Such vocabulary mixups are rampant, Wack said.
“Part of the market problem today is everyone does not have the same understanding of words and things,” he said. When standards do emerge, they’ll need to have glossaries or dictionaries embedded.
Banks don’t need to build their own blockchains.
Scores of banks have been gravitating toward the Hyperledger Project, the Enterprise Ethereum Alliance and R3. It’s becoming apparent banks don’t have to go it alone, though a few have.
That would be akin to every bank developing its own version of the Windows operating system, Bennett said.
“There is no competitive differentiation in having your own version of Windows, because the competitive differentiation is in the applications you build on top,” she said.
Some banks that have built their own blockchain technology “have realized they were doing themselves a disservice,” she said.
For instance, because Ethereum is new and untested, deploying it on its own would leave a bank "with a massive maintenance headache," Bennett said. "It’s hugely complex, it’s full of bugs.”
The bank would have to fix all its own bugs as well as keep up with revisions to the underlying Ethereum code.
Experimentation is great, but proofs of concept don’t mean much.
Proofs of concept abound, including the Monetary Authority of Singapore’s recent announcement it has digitally represented currency on a blockchain.
But Steve Wilson, principal analyst at Constellation Research, for one, is wary.
“Frankly, so what?” Wilson said. “I can represent anything I want on the blockchain, it’s just code.”
He offered a car analogy. “You can take a finely tuned Ferrari, open the hood and fry eggs on the engine block. If anybody says that’s a proof of concept for cooking food on a car engine, fine, but that’s a pretty crazy concept,” he said. “I ask people, what concept are you proving with all these POCs?”
Also, a blockchain deployment with two or three counterparties is not a fully-fledged blockchain network, Bennett said.
“That’s where a lot of the difficulties lie. It’s comparatively easy to get something working between two or three parties because then it’s easy to agree,” Bennett said. “What we still don’t have is anything that scales and more importantly, anything where you can be sure other companies will accept the same processes or data standards.”
Standards and agreements need to be worked out, but there doesn’t need to be one set of standards.
“Industry standards, structures and legal agreements have to be sorted out first before we can get the technology to work in the financial space,” Skinner said. “How are you going to build the agreements around who has access and how?”
But there may end up being many financial blockchain standards.
“We could easily see, 10 years from now, five different blockchain currency systems interoperating,” Skinner said “And there’s no reason to have only one. We don’t have one internet standard.”
Blockchain technology may never replace people and processes.
An early hope for blockchain technology was that it could cut out middlemen from certain processes.
“In banking, people and process is super expensive: you have auditors, managers, chains of authority and so much paperwork,” Wilson said.
But people are needed to deal with errors and disputes.
“Without any people or process you’re on your own,” Wilson said. “There’s that famous case where someone fat-fingered 1,000 bitcoin instead of 10 bitcoin and they couldn’t reverse the transaction, it was only because the receiver was good enough to pay them back that everybody walked away happy. In pure cryptocurrency, that’s great. In institutional banking, you have to scratch your head and say, to what extent can you get rid of people and process?”
On the other hand, blockchain-like technology will help banks achieve greater efficiency.
“We will get to a point in a couple of years where there’s a new way for banks to settle ledgers,” Wilson said. “They just won’t look like the blockchain and it would be dishonest to call it a blockchain. It’s going to be a distributed ledger. It’s probably going to be private, it may be proprietary and I think that’s a good thing.”
Not all so-called distributed ledgers are really distributed.
IBM-affiliated blockchains are housed in IBM data centers, Wilson pointed out.
“We used [distributed] in blockchain because Nakamoto was trying to get away from institutions and distributed was a political objective in and of itself,” he said. “It was a core objective and achievement of blockchain to be distributed, because it was thumbing its nose at the Fed. Distributed in actual banking is academic. We want redundancy and we want systems that don’t go down and you want your communication to go from London and New York in a redundant way.”
Banks don’t really want many of the components of a blockchain.
Immutability and transparency are among the most touted features of blockchain. Those things likely aren’t big draws for banks.
Given the “fat finger” example above, the ability to fix or change something is important to banks, Wilson said.
Also, transparency raises lots of issues.
“A pure blockchain architecture is completely and utterly untenable in that environment because any data that is stored on a blockchain is immediately visible to everybody,” Bennett said. “The moment you’re exposing personal data to parties that are not supposed to see it, it’s clearly a bad thing.”
Exposing data that is commercially confidential, such as trading strategies, client names and prices is also problematic. Regulators who saw banks were sharing prices could accuse them of being a cartel.
“If you want scale, confidentiality, control, authentication, permissioning, etc., you’re already talking about governance mechanisms where only the counterparties see the details of a transaction,” something blockchain architecture doesn’t support, Bennett said.
Blockchain vendors speak of keeping some information “off chain,” but again, that defeats some of the purpose of using a blockchain in the first place.
“Where I’ve seen a lot of these projects go wrong is they start with the technology and a use case and slam them together at 100 miles per hour,” Bennett said. “You’re running into a brick wall because you realize the technology you’ve chosen doesn’t meet the requirements of the use case.”
Blockchain technology may inspire change, rather than be the change.
Bank technology consortium R3 recently shifted from calling its partially built platform a blockchain to “blockchain inspired.” This may become a theme as banks and their vendors continue to veer further from the original principles of the blockchain.
“For many of the use cases banks are working on, the main contribution of blockchain will be as a catalyst to think differently about how to deliver this system,” Bennett said.
“We’re going to look back and say, blockchain was tremendously inspirational, it opened people’s eyes to new ways of sharing data and new ways of reaching consensus.” Wilson said.
Editor at Large Penny Crosman welcomes feedback at firstname.lastname@example.org.