How to decide whether to deploy blockchain

While blockchain may have moved beyond the proof-of-concept phase this year and into limited production systems, that doesn't mean companies watching from the sidelines should plow ahead with their own deployments.

But neither can they afford to sit idly by.

"You can't catch up on innovation. If you wait until things have settled down, it may be too late," said Forrester Research principal analyst Martha Bennett.

Speaking at Forrester's New Tech & Innovation Conference, Bennett said public and private organizations must first determine what business processes blockchain distributed ledger technology (DLT) can address – and those to which it cannot be applied.

"Nothing is being revolutionized today from an enterprise perspective," Bennett said. "Quite frankly, it is a wild west out there. When you make comparisons with the early internet, which a lot of people do, there are indeed some parallels in that we really don't know where this is going to go yet. But that doesn't mean we shouldn't engage with it today."

Sorting out use cases

To determine if there is a use case, enterprises must first understand that while many existing technologies, such as relational databases, can already address most transactional business needs, they cannot match blockchain's key attribute: collaboration.

IDG

"Distributed ledgers are a team sport. It's about data you can trust to the highest degree possible and it's about sharing," she said.

Businesses should ask themselves whether multiple business units or other industry participants have the same issue or a related issue, such as transaction reconciliation problems, and what opportunity they can capture through blockchain.

For example, we.trade, a consortium of nine competing European banks, deployed a distributed ledger over which trade finance can be transacted for small- to medium-sized businesses. The DLT enabled the banks to continue to compete and it addressed a common issue: simplifying domestic and cross-border finance.

Bennett offered up a "check list" enterprises should consider when determining whether to deploy blockchain, including:

• When multiple parties need the same data and the ability to write to the same data store;
• When all parties need an assurance that data is valid, and hasn't been tampered with;
• When a current system is error-prone, too complex, too unreliable or full of friction points;
• When there are good reasons not to have a single centralized system, such as a relational database.

The Fed takes a look

Paul Brassill, vice president of IT for the Federal Reserve Bank of Boston, said his organization, while typically cautious, has been struck by the "hockey stick" effect of technology adoption over the past two decades, especially blockchain. The Fed first began exploring the use of Ethereum blockchain technology in 2016, then moved onto Hyperledger Fabric; it has invested significant resources determining how blockchain may affect, not only the national banking industry, but worldwide financial services.

"If the private sector can do it better than the Fed, then we need to do it better," Brassill said. "We started by uploading Ubuntu Linux and installed very simple smart contracts. Our learning along the way included the fact that our developers didn't know anything about this technology. We had to teach ourselves a lot of this."

The Fed's developers watched Youtube videos on how to use blockchain; one that was particularly useful, Brassill said, is an IBM-produced video about developing a blockchain with Hyperledger Composer.

The Fed's developers used Amazon Web Services to spin up a Linux virtual machine and download Ethereum Fabric.

IDG

A "pivotal moment" came when the Fed sent IT staff to Europe to talk to their counterparts at the Bank of England, which was exploring moving the nation's interbank settlement network onto Hyperledger because it is a "permissioned" blockchain, not an open DLT on which cryptocurrencies such as bitcoin are based.

"We realized a banking network of the future needs to have membership services; it needs to have more [encrypted] transactions, so we abandoned our Ethereum approach and started going in the Hyperledger approach," Brassill said.

The Fed's three blockchain pilots

The Federal Reserve Bank of Boston built three blockchain proof-of-concepts, first to test it as a general electronic ledger for the more than $6 billion it manages in reserve for the region's banks.

A second proof of concept tested what an audit or supervisory node in a distributed ledger might look like if it were part of a massive, digital national banking network.

"Would it do auditing or look for anti-money laundering challenges, or would it look for unusual fraud behavior and how would you digitize what is today a very human activity," he said.

IDG

"We have to build a spider web of connectivity between all of these banks where we're watching transactions over some massive blockchain environment and we're going to be in the middle of it," Brassill continued. "We're going to be ordering those transactions; we may be arbiter of the membership services deciding who gets to be involved in this network, who issues certificates for it. This is a massive challenge, but I think it's one that's necessary to see FinTech and DLT really impacting the financial market."

A third blockchain PoC the Fed is building will run a non-critical HR application – one of its HR employee appreciation functions – but it will be run 24/7 to determine what problems might arise from constant use, such as scaling issues.

"Will we have storage issues..., smart contract development [challenges], what are the cyber risks?" Brassill said. "So, before we build out this mission-critical general ledger platform in the future, we first should get our arms around something very modest."

The Fed is aiming to roll out a production version of that blockchain-based HR app over the next year or so.

The agency is also examining how DLT could disrupt processes around the nation and the world, such as through micropayments – small online payments most often between consumers and retailers. And the Fed is looking at how blockchain could eventually affect large retail and wholesale payment platforms, such as the Automated Clearing House electronic network and Fedwire, which is used for large wholesale payments between banks.

"What if those platforms are running on distributed ledger technology? What are the cyber risks? How fast do they run? We need to understand that before we're disturbed," Brassill said. "Then there's this whole idea of fiat cryptocurrencies.... The Fed needs to understand where this is going. What do ICOs mean for the economy? What do all of these 2,500 cryptocurrencies mean? By extension, what does this mean for the global eco-structure?"

The Federal Reserve Bank of Boston also has 300 examiners whose job it is to ensure the region's banks are liquid and stable.

"Picture a decade from now if some of those banks are running their credit system on blockchain; what might our examiners need to understand to better evaluate those companies?" Brassill said.

Using blockchain for larger networks

The Fed has been trying to determine where it stands in comparison to the world's other central banks in deploying blockchain because in the future there may be a global blockchain network through which banks transfer both traditional fiat and cryptocurrencies. It plans to release a white paper by the end of the year detailing its experiences rolling out blockchain PoCs – not only from a governmental but a business point of view, Brassill said.

Over the next 12 to 18 months, the Boston Fed will begin tying into its PoC other Federal Reserve banks with the goal of creating a sandbox to test applications.

Because enterprise or "permissioned" blockchain's are distributed, anyone allowed into the electronic ledger by a central authority can potentially see every immutable data entry; that's extremely useful for business processes such as supply chain tracking or cross-border payment and settlement. Smart contract technology can also control who on a ledger gets to see what.

But, Bennett cautioned, smart contract technology belies its name: it is neither smart nor a legal contract. Smart contracts are business automation constructs (software coded to enforce predetermined rules) with the ability to achieve disintermediation of what are often manual processes.

And even when a business has successfully tested blockchain networks, even multiple times, it still doesn't mean the technology is ready for production because it still may not scale; it will need to integrate with existing business systems; and it will need to meet regulatory and compliance requirements that in many industries have yet to be determined.

"That's one of the reasons we're not going to see large-scale adoption immediately," she said. "Anybody who talks about DLT revolutionizing this industry, this process, this whatever – nothing is being revolutionized today from an enterprise perspective. What companies are doing is really looking at what they need to do today in order to do things differently in the future."

For example, last year, Northern Trust Corp. launched a commercial blockchain network based on Hyperledger Fabric for managing the administration of a private equity fund. The DLT network enabled real-time document exchange between all involved parties, and smart contract technology streamlined the approval process.

"They reduced the time taken to get together all the paperwork required for a private equity transaction from three weeks to three days," Bennett said, adding Northern Trust also allowed an auditor onto the blockchain in order to inspect the transactions in near real time, as opposed to the more traditional approach of performing periodic audits of long-completed transactions.

Why companies might avoid blockchain

Companies considering deploying blockchain should ask themselves whether there are good reasons not to have a single centralized system, because DLT introduces complexity and, at least today, still introduces risk because of a lack of maturity.

"Ultimately, only you know your organization and your industry well enough to be able to say this is where the shoe hurts and this is the process that's broken. What problem are you trying to solve with this?" Bennett said.

She cautioned that while there are parallels between the internet's early days and blockchain technology, in that both had the potential to become a ubiquitous communication vehicle, the glaring difference is blockchain has not undergone rigorous academic research.

"Typically, what you get, whether its artificial intelligence, or augmented reality, nanotechnology, or quantum computing is they tend to start in the research labs...then they go into a commercialization phase for use in the real world, and then they go mainstream," Bennett said. "We've got networks running that are inviting commercial participation which haven't been undergoing any form of rigorous academic research."

What many of the public and private organizations that have deployed blockchain have learned that distributed ledger technology is far from fully baked and still faces hurdles that could take years, if not a decade, to work out.

For example, there are still a myriad of competing specifications and blockchain iterations, so it's not known whether blockchain acceptance will be driven by standards, industry initiatives or de facto adoption.

"We'll have to wait and see," Bennett said.