Over the past six months the fintech world and financial industry as a whole have gone agog with the term "blockchain." We can see that trend depicted via Google search results in the chart at right.
Why the sudden interest in "blockchains" and "distributed ledgers"? After all, much of the technology and ideas used by Bitcoin, Ethereum and even Git, the open-source system for managing software updates, have been around for more than a decade. Public-key cryptography has existed since the mid-70s. Consensus algorithms such as PBFT and DLS used in distributed computing and distributed databases have likewise been under development and in production environments for decades.
While all the individual elements that comprise the Bitcoin blockchain have been around since 2001, it took until Satoshi's 2008 white paper to demonstrate how these individual pieces could be cobbled together to work as one to fulfill certain tasks. Specifically, the paper described a new type of data structure that was tamper-evident and economically costly for any one entity to unilaterally revise.
Careful study of cryptocurrency systems such as Bitcoin showed that under certain security assumptions, it is possible to verify and authenticate when financial information has been transferred and updated. Due to how it was configured, the specific applications of Bitcoin's blockchain itself resulted in a niche set of use cases, namely pseudonymous interactions that needed censorship resistance on a public network.
What happens if you reconfigure those same elements or substitute those elements with other cogs and software libraries? Different types of utility and applications arise from this mix-and-match process, all of which have pros and cons depending on the operating environment. One set of gears includes a family of tools used to enhance privacy of participants and is found in projects such as Enigma, zk-SNARKs, confidential transactions and a handful of others. Each of these has trade-offs depending on how they are integrated with the host network. And in some cases, the resulting system itself caters to use cases that are gated and not censorship-resistant.
Some Bitcoin enthusiasts assert that non-Bitcoin blockchains are all the same and that "permissioned" blockchains – or blockchains that gate transaction validation – are not much different than relational databases mixed with snake oil.
The problem with this broad generalization is that it does not distinguish between what each project, startup and effort in the overall space is trying to do. It lumps them all together without bothering to distinguish the potential customers and problem sets these projects are addressing.
For instance, most traditional financial institutions must comply with a variety of regulations that effectively rule out use cases that involve censorship resistance. In almost all cases, regulated entities must "know your validator." Therefore, since all transaction validators participants are known and legally accountable to off-chain third parties such as courts, the intensive proof-of-work mining process used by many public blockchains becomes unnecessary and wasteful.
But wasn't the sole value proposition of Bitcoin censorship resistance? If you remove that, shouldn't you just use an existing database for any and all interactions?
The problem is that while many financial institutions do share information with one another through a variety of APIs and "wires," they still have to rely on external third parties and expensive reconciliation systems to transfer and reconcile assets between one another. There is no such thing as an off-the-shelf, production-ready, cryptographically signed shared ledger specifically designed for large-scale enterprise usage. At least, not yet.
But what if institutions could use elements of a blockchain, such as public/private key cryptography to sign transactions and atomically settle registered assets without a third party? What if reconciliation and auditability could be cryptographically proven within the data structure itself, thereby reducing the need for some of the existing back-office services? What if financial institutions could share one common ledger that was specifically designed for their regulated operating environments? A well-designed and built shared ledger could make all that possible. Conceivably, as one report from Santander predicts, up to $20 billion could be saved per year by using some of this non-Bitcoin-specific distributed ledger and blockchain technology.
But why not Bitcoin or Ethereum themselves? Why avoid public networks? Is it just that "a blockchain" – the brand itself – has a certain je ne sais quoi about it?
Many Bitcoin enthusiasts proclaim that since the Bitcoin network already exists today, there is little reason to go through the effort of reinventing the wheel. The reality is that the Bitcoin blockchain – which is just one of a few hundred blockchains – cannot currently provide a secure on-chain solution for the settlement of off-chain registered assets. Nor does it have the ability to quickly upgrade to provide such utility or capacity. As Shane Leonard, co-founder of StockFlare recently pointed out:
Remember, Bitcoin is just one "instrument." There are 45,000 companies listed on stock markets alone. There are 400,000 legal entities that have a "Global Legal Entity Identifier." Reuters have a database of over 35 million instruments. Imagine what sort of blockchain would be needed to manage all this data and these participants.
The Bitcoin blockchain – which is just one of a few hundred blockchains – cannot as it stands provide a secure on-chain solution for the settlement of off-chain registered assets. In fact, as illustrated by the current block size debate, approximately $500,000 per month has been spent simply debating what the block size on the Bitcoin blockchain should be. It is a collective action problem that arises due to the very nature of how its cogs were purposefully arranged seven years ago.
It is a truism to state that there are many different types of networks and data structures that provide different types of utility. There is a time, a place and a customer base for pseudonymous cryptocurrency systems. But for now, the global financial services markets demand far more than the Bitcoin blockchain can handle.
For the average reader of American Banker, those working for and representing globally regulated financial institutions, certain types of shared ledgers, with legally accountable validators and clear terms of service, could likely provide unique utility to your organizations. Whether the end product is called "a blockchain" or not is a No True Scotsman fallacy and one that distracts your firm from making economic, cost-saving decisions.
Tim Swanson is director of market research for R3CEV and is based in the San Francisco Bay Area. Follow him on Twitter @ofnumbers.