The topic of blockchain drew the most discussion in Fintech for 2015, not THE bitcoin blockchain but just the general concept of a shared distributed public ledger. Financial institutions (FIs) have varied views about what blockchain technology can mean to them. For some FIs, blockchain has warranted serious exploration and investment. According to the Aite Group, financial institutions spent an estimated $75 million on the technology in 2015.
For many FIs, however, blockchain’s definition and function remains elusive. Recently, I had a discussion with a bank executive who wanted to know how they could use blockchain technology to develop a centralized, private, and encrypted ledger. He understood the blockchain as a new technology that his bank could mold and, in essence, use to enhance the current banking models. Blockchain defies centralization, but may be private and encrypted in some fashion.
It’s design as a distributed ledger, as opposed to a centralized ledger, makes blockchain especially intriguing. (Distributed means that every participant in the blockchain has the same exact version of the ledger all the time.) Mike Gault, Founder and CEO of Guardtime writing in re/code puts it this way:
Imagine that you’re walking down a crowded city street, and a piano falls from the sky. As dozens of people turn to watch, the piano crashes down right in the middle of the street.
Then, without a second to lose, every person who witnessed the event is strapped to a lie detector and recounts exactly what they saw. They all tell precisely the same story, down to the letter.
Is there any doubt that the piano fell from the sky?
Because everyone participating in a blockchain has the same version of the ledger, participants do not need to refer to a single, central copy, which the current models require. Improved efficiencies result from having a distributed ledger. For example, in capital markets settlement of a stock transaction takes three days. With a blockchain design, such process could take place much faster. The distributed nature of a blockchain would eliminate the need to have all the “witnesses” corroborate their accounts of the piano accident. Such design would eliminate disputes, making processes more efficient.
While a blockchain can be private (not open to the world to see), it has to remain public to all the participants in a blockchain. A blockchain’s distributed nature can only remain if it stays transparent. If FIs remove the transparency of a blockchain by making it visible only to particular participants, such as to the FI, they just replicate the existing models. This approach misses the point of a blockchain.
Blockchain purists argue that the essence of a blockchain’s distribution lies in its total transparency, and that its recorded events must remain entirely visible to everyone, whether they participate in part of that specific blockchain or not. While they have a point, limited access can enforce a level of privacy. For example, the bitcoin blockchain is theoretically public, yet if someone doesn’t participate in bitcoin, they don’t have access to the ledger.
The contents of a blockchain can be encrypted but if the deciphering keys remain under central control, once again they are replicating existing models. For a blockchain to remain functional as a distributed ledger, all the participants in a blockchain need to be able to decipher its contents. Naturally, FIs have a desire for encryption in order to ensure that only participants in a blockchain have access, but that must not interfere with what makes a distributed ledger attractive.
In the case of the piano accident, a blockchain allows for all witnesses to have a complete and identical account of the incident, while limiting it only to participants, thereby assuring privacy. Applying privacy controls or encryption to the accounts of the accident limits the ability to ensure identical stories, and would make the blockchain design ineffective.
As FIs consider the use of blockchain technology, they should be careful not to bring the current models into the discussion. FIs should start with a goal in mind and determine how a blockchain might help them achieve that goal.
FIs looking for improvements in processing times need to define new processes that remove the inefficiencies of current models. For example, Nasdaq Linq (a blockchain for capital markets) promises significant improvements due to shorter timelines. The latest results show that trade clearing and settlement time is being reduced from three days to less than 10 minutes. Such time reduction also reduces risk exposure (the possibility that funds won’t be available to cover the purchase by the time of settlement) by 99%.
FIs also need to identify the participants in these new models and include them in the design process.There are at least two major industry groups that exemplify this approach. The partnership between R3, a blockchain startup, and 30 or so major banks including Citi, Bank of America, and HSBC, seek to build a real time fund transfer network. A consortium headed by Digital Assets Holding’s Hyperledger, IBM, Cisco, Intel, the London Stock Exchange, and several banks is working on various use cases for blockchain.
As FIs look at these new models which include some level of openness in comparison to current models, they need to come to terms with balancing transparency and security. Also, they must consider what value FIs can bring to a blockchain model as well as any revenues they may need to replace. As the industry moves forward in adopting blockchain technology, FIs will need to answer these questions.
The question of future regulations still remain, and warrants a separate discussion.
A version of this post was published by Irish Tech News under the title “Hey you! Leave my blockchain alone. Why blockchain doesn’t need centralising” on January 29th, 2016.