This post first appeared on Medium
Are banks really joining the blockchain bandwagon?
If you have the misfortune to work in “fintech” you’ll have spotted a trend in the last 6 months, the blockchain and banks. Barely a week goes by and another investment bank launches an innovation hub, makes a speech about the blockchain or joins an initiative.
For example, Oliver Bussmann, Chief Information Officer at UBS: > I believe — and this is my personal view — that blockchain technology will not only change the way we do payments but it will change the whole trading and settlement topic.
Looking at Google Trends for “blockchain”, you’d be forgiven for sitting up.
Statistics can lie
How easy it is to manipulate data. Compare “bitcoin” to “blockchain” on Google Trends and immediately the reality is clear. The concept of the blockchain has a long way before it enters popular culture.
Looking at word-count in the New York Times and Wall Street Journal, we see a similar picture.
Forget the manipulation, do banks really care?
Yes, they do. But not in the idealistic way the supporters of Bitcoin do.
The early supporters of Bitcoin imagined a way to replace banks, the trusted-third party in most transactions, with the blockchain a near incorruptible database of ownership.
In contrast, the banks see the potential of a shared database to massively cut their overheads, not as a way to eliminate their role in the financial services industry. A shared database of transactions & ownership could eliminate significant settlement costs, back-office costs, increase the speed of transactions, reduce errors, and potentially reduce their capital requirements. For banks, the blockchain has the allure of a veritable nirvana of lower capital requirements, lower costs and potentially higher revenues.
The stock market example
This graphic shows the complexity of selling shares on a stock market.
The complexity of the a stock market trade is to ensure no failed transactions, to ensure you own what you sell and get paid for it by the buyer. In that way, it’s like Bitcoin’s blockchain ensuring no double-selling.
So in theory, a blockchain for the stock market could let the seller and buy meet each-other without the need for brokers, custodians, a stock exchange, a securities depository, or share registrar. I sell, you buy, transaction recorded. No “settlement risk”.
But sadly it’s all a pipe dream, at the moment. An exciting one, but with enormous obstacles, if the experience of Bitcoin is to be learned from.
A cautious approach
All major financial services firms are regulated. Banks are almost utilities. They value their brands and the trust clients place in them. Any technology that could undermine trust in a financial institution is treated cautiously. As Walter Bagehot, one of the founding editors of the Economist newspaper noted:
Every banker knows that if he has to prove he is worthy of credit, in fact his credit is gone.
Which is leading the financial services industry to consider a more controlled approach to building shared databases and explains their willingness to join consortia such as R3, who is seeking to establish open standards and open protocols. But as Dr. Lee Braine at Barclays points out:
Whether you are looking at permissioned or permissionless ledgers, it’s obviously necessary to ensure security, reliability, performance, etc.
These institutions store trillions of dollars of wealth for their clients. They process trillions of dollars of transactions daily. They don’t play loose or fast. Not if they want to remain in business.
But why do banks exist at all?
Money serves two purposes. It is a store of wealth. It facilitates transactions.
Its whole existence relies on property rights. Its bedrock is trust. When a society stops trusting its money, chaos, economic depression and sometimes violence ensues.
Before the advent of banks, you had “gold coins” both as a store of wealth and to transaction. Money was a “bearer instrument”. Possession equalled ownership.
But storing your wealth is problematic with “bearer instruments”. Why? Theft or loss. Thus the invention of banks as an efficient way to store your wealth.
Furthermore to trade with someone, it’s painful for the buyer to have to always take the “bearer instrument” with them, hand it over for the goods, only to see the seller go back to the bank to lodge the “bearer instrument” back with the bank.
Thus banks have played a valuable role for centuries both as a place to store wealth and as facilitators of transactions (i.e. trade). But “bearer instruments” have their limitations. As anonymous instruments, they can be used criminally and to avoid taxes. As a physical product they can be lost, stolen and could not be transferred electronically.
The inefficiency of “bear instruments” led to the invention of cheques, promissory notes, letters-of-credit, direct debits, bank transfers, etc.
These solutions can be referred to as “registered instruments” where there is a register of ownership. Whoever’s name is on the register, is the owner. The move to “registered instruments” has lead to a massive increase in trade, which increases wealth, which leads to innovation, which leads to more trade, and so on.
In economic terms, think about the “Quantity Theory of Money”:
PQ = MV
The size of an economy is the price of all good sold (P) times the quantity sold (Q). Since we pay for them, if there is a fix amount of money (M), then the speed the money circulates (V for velocity) impacts the size of the economy. The move from bearer to registered instruments positively impacted velocity.
But with “registered instruments” there is a database to manage. They are managed by third-parties, banks, clearing houses, stock exchanges, etc. Trust is required to ensure a well functioning financial services industry and to facilitate the inner workings of the economy and society. Regulators play a key role in ensuring that trust is maintained.
Hotblooded capitalists that we are, we accept the costs of regulation, given the massive benefit of our current financial system to the economy & society as a whole, relative to a system based on “bearer instruments”.
In effect, the mechanism of our financial system has gone from:
Barter to Bearer to Registered Instruments.
So you’d be forgiven for doubting the merits of going back to a “bearer instrument” based financial services system. If that is what blockchain technology leads to.
Going round in circles
No one “owns” the financial services system. There is no dominant player. Banks have shareholders who expect to see the banks make a return on capital, as a result of the services they provide to their customers. The primary services banks provide to clients are storing value and enabling transactions.
The Bitcoin eco-system is no different. Just a lot smaller in scale than the rest of the financial service system.
No one “owns” or controls it. There are miners who maintain the blockchain. They have to make a return on the capital they have invested. They facilitate the storage of value and enable transactions, albeit at a fraction of the scale of the rest of the financial services industry. And just like the rest of the industry, it functions because there is trust. Though the trust in a third-party entity is replaced with trust in the network of miners who maintain the distributed ledger.
What lessons can banks learn from Bitcoin?
Let’s look at four issues: 0. Trust & intermediaries 0. Scalability 0. Operational costs 0. Speed
(1) With Bitcoin, trust has not disappeared as a vital consideration. In fact Bitcoin replaces the need for trust in the financial services industry with the need to trust the functioning of the blockchain. Yes, there are both theoretical and practical reasons why users have a high degree of trust in the blockchain, but there is still a need to trust it. Today, there is a high degree of concentration in the Bitcoin miners who secure the blockchain, with 5 entities controlling over 50% of the network. This concentration may already be a cause for concern.
Similarly, there are still intermediaries, just like in traditional financial services. Many investors store their Bitcoins in wallets at third parties. It’s similar to having a brokerage account or bank account. And you need to trust the operator of these wallets. The failure of Mt.Gox highlights that the issue of trust is as relevant for Bitcoin as in other parts of the financial services industry.
(2) Bitcoin faces an issue of scalability. Whereas Visa can process over 56,000 transactions a second, in it’s current form, only 7 Bitcoin transactions can be processed a second . Back to Dr Lee Braine of Barclays: Blockchain-inspired solutions such as shared ledgers and smart contracts should aim to meet the enterprise-scale architectural non-functional requirements.
The scalability issue has caused some infighting in the Bitcoin community, with many supporting a “fork” in the technology to allow it scale. And the debate has got so heated that one of the leading developers, Mike Hearn recently wrote:
The Bitcoin Core project has shown it cannot reform and so it must be abandoned.
(3) Worse still, the resources required to manage the blockchain are going exponential. The more trading the larger it gets. Yes, storage costs typically decline exponentially too, but they are a cost nonetheless.
Remember, Bitcoin is just 1 “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.
Perversely for a “bearer instrument”, the blockchain has an operational cost. In contrast, for central banks, printing currency is hugely profitable with little ongoing operational costs. The cash just circulates.
In the case of Bitcoin, the miners are paid to maintain the blockchain, they effectively “print money” like a central bank, receiving 25 Bitcoins for each successful update of the Blockchain. But these “subsidies” aren’t indefinite. Then the miners will need to receive transactions fees to cover their costs.
(4) The blockchain is updated every 10 minutes. Even if the system was changed, most engineers recognise that the speed of any distributed-ledger like Bitcoins will under-perform a relational database in terms of speed. Yes it may be more trust-worthy, but the speed at which it is updated is unlikely to ever match a simple database.
Given the limitations in terms of speed and scale, it’s not surprising that there has been an explosion in the alternatives to Bitcoin’s blockchain. As mentioned earlier, most major financial institutions are working with R3. Ripple is working on the Interledger Protocol to facilitate settlement of foreign currency payments, instantly at low cost. But the list goes on: Blockstream’s Sidechains, Hyperledger, NXT, Ethereum, to name a few of the notable ones.
So is there an obvious application for banks?
Again, the answer is yes, the clearing and settlement systems.
As Robert Sams, the CEO of Clearmatics pointed out:
It takes days to settle trades in book entry assets. This fact is only puzzling to those labouring under the mistaken assumption that custody accounting in the financial system is somehow centralised. It’s not.
Records are distributed throughout the system by thousands of different institutions, each maintaining their own siloed accounts and constantly reconciling against each other to come to agreement on the global state of who-owns-what, or who-owes-what-to-whom. It is, in a sense, a form of distributed consensus: consensus-by-reconciliation. And consensus-by-reconciliation is very slow, expensive, and hard to automate.
It is this technological infrastructure of consensus-by-reconciliation that the bankers, quite rightly, see being replaced by distributed, shared ledgers.
So Oliver Bussmann at UBS is in agreement on this, trading and settlement systems are ripe for innovation.
But as Jamie Dimon of JPMorgan points out, there are players like the DTCC who already manage the ownership data. Imagine, the DTCC “stores” $37 trillion of instruments and manages $1.6 quadrillion of trades a year. In contrast, Bitcoin has a value of just $4bn with daily volume of $80m ($29bn annualised). Let’s have “fun” and put that in a table:
Therefore if blockchain technology is to have a place in mainstream finance, there is a need for an enterprise level solution, where security is strong, data protection laws are respected and lower costs don’t come at the expense of performance.
Will consumers benefit?
Of course, consumers and companies will benefit, if banks succeed in rewiring either/ both the payment networks and the custody/settlement systems.
More efficient banks, faster transaction times, less errors, etc are all a positive for society. Though employee numbers in the back-offices of banks will be hit hard. Banks that innovate will be rewarded with higher profit margins and returns of capital should rise.
With an increase in the speed of transactions, the blockchain innovation may have an ATM moment. i.e. massively increase the access to money, enable increase trade and improve financial access. In economic theory, the ATM increased the velocity of money, V in the Quantity Theory of Money.
Paul Volcker, a former Chairman of the Federal Reserve joked that:
The only useful thing banks have invented in the last 20 years is the ATM.
So perhaps, we are about to see the next truly useful innovation by banks. With blockchain technology driving a step up in global trade, wealth creation and innovation.
But make no mistake about the lessons learnt from Bitcoin, it is like a petri dish experiment for the financial services industry. And the experiment highlights just how hard its going to be.