Yes, but what’s new with distributed ledgers?

This short post is inspired by a conversation I had recently with a couple of finance professors from top business schools who had some questions about blockchains.

Prof A explained that he had heard all the fuss about blockchains but was unsure whether it was revolutionary or evolutionary (I think the word disruptive was also used). I have written about disruption in Fintech and the Evolutionary vs Revolutionary aspects of distributed ledgers before (hint: it depends, it’s both, and yes, perhaps).

Then he asked, “Yes, but is there anything new?”

I paused: it was a good question.

There isn’t anything technologically new about blockchains, or distributed ledgers in general. Digital signatures? Seen them before. Hashing? Ditto. Databases? Old-school. Peer-to-peer? We get it. Consensus mechanisms? Caught ’em all. Even proof of work (gentle reminder: not all distributed ledgers need proof of work)? So 90s. No, there are no breakthroughs of that sort.

So what’s new?  The way the building blocks are put together create new uses.

With cryptocurrencies the novelty is that for the first time you can have a digital token of value not backed by an entity of any kind, that can be transferred across the internet from one pseudonymous account to another, knowing that the system is generally ‘honest’, without a central party doing the account keeping, and you open accounts by rolling dice (metaphorically) instead of by asking a 3rd party if they will kindly assign you an account number. In other words, unstoppable electronic cash.

Where is that helpful?  Anywhere you might want to use almost anonymous electronic cash, but bear in mind the value of the cash fluctuates, and to turn the electronic cash into local currency you may need to identify yourself.

What about distributed ledgers? Here the novelty is that for the first time you can easily create a system where multiple entities have shared control over the evolution of data.

I should briefly define “evolution of data”. For example if a database is keeping track of the ownership of a thing, and at first I own the thing, and then I give it to you, so now you own the thing, that’s data evolution: the data saying who owns it has evolved. The database would have two rows: the first saying I owned it, and the second saying you own it now so please ignore the first.

Where is this helpful?  When you have multiple parties who want to have an equivalent say in the evolution of data that is important to them, and they would like to have shared control over that data. Perhaps historically they used a 3rd party or you relied on a series of reconciliation steps to “just check” that the other party thinks the same as them. And perhaps a 3rd party controller of the data doesn’t make sense due to cost, privacy, or political reasons. In finance, that’s almost everywhere. (The “disintermediation of 3rd parties” narrative is also widely misrepresented, but that’s for another post).

So: although the building blocks are old, the way they are put together enables some new uses, from anonymous(ish) cash that attempts to make banks irrelevant, through to reconciliation technology that is sorely needed by industry.

Sure, if you’re being pig headed you can argue “Yes but I can build a system of unstoppable cash by having some databases, joining them up in a peer to peer way, adding some digital signatures for verification, making people repeatedly hash to slow the system down, etc etc” but you will have designed bitcoin. Likewise you can say “Yes, but I can build a system where participants have shared control of data by getting some databases, connecting them together with APIs, using some digital signatures, creating a consensus mechanism to resolve race or double spend conditions, etc etc” but you will have designed distributed ledgers.

So the question isn’t about what building blocks are new, it’s about what you’re building with the blocks.  After all, Uber is just phones, internet, and taxis.