I was reading the Philippa Ryan’s review of two Data61 (CSIRO) reports (here) on the slow up-take of blockchain in SmartCompany (here) and it made me want to revisit my review from last year (here), which was pretty negative.
It’s gratifying to have them echo much of what I wrote, but they raise the issue of smart contracts. Is this the ‘killer app’ that could make it all worthwhile?
Smart Contracts – the Killer App for Blockchain?
The automatic nature of smart contracts could have huge potential. Particularly where there are issues of counterparties living up to their contractual obligations, for example:
- Shipping goods to a customer, hoping to be paid after they’re received
- Payments dependent on inspections, potentially by a 3rd party
- Delayed payments for contracts that extend over time
In each case there is a contract that requires one party to live up to their part in the hope that the other party will live up to theirs at some later date. These are notorious for problems where the second party may find an excuse (or not) to delay or dispute payment.
If payment contracts like this can be automatically enforced, it would give great comfort to those who need to live up to their end of the bargain first. This is a huge problem locally, but even moreso for international trade. Different regions have culturally different approaches to the value and enforcement contracts, which has caused incalculable problems of late.
How good would it be to know that you’re 100% guaranteed to be paid for your goods once the freight company logs their Proof of Delivery? Or that your retailer will pay you automatically for each item they sell, without interference? Or your bonus will be calculated automatically once the audited accounts are lodged?
Clearly a game-changer – but is Blockchain necessary?
Blockchain technology is a truly elegant solution to a lot of computationally interesting problems. But (see my previous article, reproduced below) a lot of those solutions are for problems we don’t really have today.
Despite millions of disputed transactions every year, I’ve never seen one that was due to a repudiation issue or a transaction ‘lost’ between banks. The systems in place are old and creaky, to be sure, but they work. They simply don’t have most of the problems that blockchains can solve.
And automatically-executing contracts could be implemented fairly simply without the blockchain underneath them.
We already have escrow agreements, banking instructions and other forms of 3rd-party enforcement of contracts. Dragging these into the digital age would be a huge step forward for everyone. But I don’t think we need to transform the entire banking infrastructure to make that happen. Sadly.
Banks and other financial intermediaries could easily implement automatic contracts, either manually or automatically, based on what they have in place.
Selling them to their customers could be another story. Like the ‘Minimum Viable Product’ of smart contracts, escrow agreements and letters of credit for international transactions could be easily revamped for the 21st century. This would give us a really interesting view of how attractive removing the human element from payments could be.
And I, for one, would love to see that!
Shameless Self-Promotion II: Looking for coaching/mentoring clients these days, so if you or someone you know has a tech, marketing, equity or M&A issue and they could use some help or a ‘2nd chair,’ I’d love to hear from you!
Here’s what I wrote last year – which I think remains true today:
Blockchain – it’s not the disruption we’re looking for.
It’s a huge focus for banks, IT providers and every would-be disruptor starting up today, but is the fintech fashion for blockchain technology going to revolutionise finance? I doubt it.
The Ethereum blockchain quickly passed 6,000 nodes and Bitcoin’s blockchain – a venerable elder in the space – has over 5,000, with buy/sell exchange rates quoted on international exchanges. Even some retailers (the hippest, of course) will accept digital currencies for ordinary purchases, both online and off.
TheDAO (a ‘Distributed Autonomous Organisation’ on the Ethereum blockchain) raised over $150M with stunning speed, buying into the promise of replacing stodgy old corporations with a fully automated businesses operating entirely on the blockchain. No employees, offices, overheads, or (maybe?) taxes involved – truly a brave new world!
But despite bold promises, big brains and millions of investment dollars this bleeding edge technology seems slow to penetrate real commercial applications. It’s already taken 8 years – ages, by internet standards – and even the largest applications are still ‘fringe,’ at best.
I for one don’t think the blockchain, at least in its current form, will ever live up to its promise.
For one thing, blockchain technology is staggeringly inefficient. It’s not bad programming (there’s very clever coding going on) but a fundamental part of the blockchain method.
Though called distributed ledger systems, blockchains are actually duplicated, not distributed. Every node maintains a full copy of every account, transaction, contract or other object put into the chain by every other node, as well as its own. By analogy, each ‘bank’ maintains the whole financial system, not just its own little part of it.
A quick search suggests there are about 15,000 banks and credit unions in the world today, each maintaining a separate ledger – often many. If joined into a single blockchain they would each presumably need to do 15,000 times as much work to maintain the combined ledger – before we even get to the fancy crypto stuff.
Great for the computer makers, but not for banks’ bottom lines.
That duplication creates another issue – visibility. The whole world (including the tax man) can see every transaction ever executed. Are we really ready for that?
There are ways to disguise your transaction history, but visibility is a fundamental feature of blockchain systems, not a bug. Each node can sum the debits and credits on your account to check that you actually have the currency you’re trying to spend.
Visibility is limited to an account number, which isn’t necessarily linked to your real-world identity – until you provide it to someone who needs to pay you, for example.
There’s some real disruption for the financial system!
What benefits are we going to get in exchange for all this effort and change?
The promise is a financial system that doesn’t rely on trust. Organisations, transactions and contracts will be executed with engineering precision instead of lawyers, courts and the whole messy system of history.
But that’s a fundamental misunderstanding of what the blockchain can do.
In the simplest case, blockchains use complex cryptography to ensure that only your personal key can transact on your account and transactions fail if there isn’t enough currency in an account. A built-in system of negotiation ensures that all nodes agree on the veracity of every transaction, and once they’re executed, transactions can’t be reneged on (at least, without the agreement of the broader system of nodes).
Blockchains are an incredibly elegant system for ensuring these basic fundamentals, but the banks have done a pretty good job with them already and a blockchain won’t prevent someone who’s stolen your key-code from accessing your account.
It will, however, stop you from accessing your account if you lose your key – ever. There’s no way to recover lost accounts, key or currency, as there is no organisation with over-arching access to the system. Lost currency is lost for good.
Fancier implementations (like Ethereum) allow for more complex operations to be controlled by the blockchain. Whole programming languages allow for complex transactions with if-then and other logical constructs to be set up as ‘contracts’ recorded on the blockchain ledger. Once the contract is committed on the blockchain, whenever the specified conditions are met, the specified transactions will occur automatically.
For example, a contract might encode that when a certain majority of a certain set of account-holders lodges their approval of a transaction, that transaction will automatically execute. TheDAO is a complex set of these constructs that operates as an investment fund, making investments based on decisions made automatically, through the votes of account-holders who’ve bought into the contract.
These contracts are notoriously difficult to program and flaws in the contracts can be irreversible. For example, if a simple contract holds $1M worth of currency until several related account-holders approve of disbursing it, when any one of those account keys is lost, the funds can never be recovered. There is no way to break the contract unless one is programmed in to start with.
Immediately after TheDAO was set up, a flaw in its constituent contracts was used to sequester a large proportion (possibly $50M) of the invested currency into another account. In this case, the overall Ethereum community decided that TheDAO was ‘too big to fail’ and worked together to remedy the situation (they hope). Such a fix is legally questionable, unlikely to happen often, and will become even more difficult as the community grows.
Blockchains substitute the problems of programmers for the problems of lawyers. Doubtless this will be an improvement, but only when blockchain programs ‘crash’ less often than traditional banks. Since there is no vendor, central bank, government or other central authority to step in, bad outcomes are final whether we like them or not.
So why is blockchain technology so popular?
For one thing, it’s truly elegant. Cryptography has been stitched together to provide a computationally beautiful solution to messy practical problems in a way that engineers (like me!) can’t help but love.
There’s also a compelling cloak-and-dagger ‘back-story.’ Satoshi Nakamoto is the nom de plume of the shadowy author(s) of the software and foundation paper behind Bitcoin in 2008, but no one knows who this actually is. Active in the community until he/she/they disappeared in 2010, the public visibility feature mentioned above shows that accounts linked to this identity still hold some 1M bitcoins, currently valued at around US$758M.
The financial incentives don’t hurt either. Early participants in mining or trading Bitcoins have built significant wealth, often with little or no risk and only a modicum of human effort – certainly an attractive proposition.
Finance Needs Disruption
I think the over-riding driver, however, is the desire to see the current financial system disrupted. Banks and financial institutions appear to many as hide-bound victims of over-regulation, greedy rent-seekers, risk-averse loan deniers or corrupt, vested political interests. Few consider them efficient operators of the economy. Those frustrations fuel a huge appetite for change.
The potential prize to a successful disruptor is huge. Bank interest rate spreads (the difference between what the bank charges for a loan, and what it pays on the deposits that fund the loan) often run around 4-6%, while peer-to-peer lending exchanges may charge only a 1% service fee. Even a small slice could easily be worth billions.
Work to be Done
Of course, that’s oversimplifying things considerably and blockchain technology is going to have to fix a few things before it can help capture that slice.
There are already efforts afoot to work on ‘sharding’ or breaking the blockchain into smaller components to distribute (rather than duplicate) some of the work. This is still in early days and there are some very complex issues required to make it work.
Ultimately, creating a hierarchy or distribution of subordinate ledgers could come to resemble the current system, with component institutions managing their own internal transactions and ‘clearing’ transactions between them. This is likely to remove or reduce some of the benefits, so a lot of careful work remains.
Another possible solution is to use specialised blockchains for specific high-value assets, like land titles, rather than every-day currency. These tend to have complexities of their own (easements on land and mineral or air rights, for example) which will inevitably take some time to work out. There may be some easier commodities to start with, like loan contracts for securitisation, but that remains to be seen.
Conclusion – Sadly, Not Yet
Overall, blockchain is a very appealing and technically elegant solution, but there’s still a lot of very complex issues to solve and limited benefit to be had over existing (albeit sub-optimal) systems.
IMHO it’s going to be several years before we see blockchain technology hitting the mainstream, and when we do, it’s likely to be very different and may come out of existing large organisations with the horsepower to drive it.
I would dearly love to see this technology, or something like it, shake up the financial world. But just now, I don’t think blockchain is the disruption we’re looking for.
But here’s to hoping I’m wrong!
This review was part of a client project I’ve just completed, so I’ve got time for more. I’d love to talk to you about how I can help you grow your business, acquire, sell raise capital or smooth out your operations. If you’re in technology, contact me to talk about how I can help.