For a discipline that is supposedly rational, finance has had its fair share of irrational crazes. For example, the dotcom bubble and the collateralized debt obligation craze that led to the global financial crisis. The latest is the Blockchain.
Most financial bubbles are based on a mistaken belief that somehow “this time it’s different” and the laws of economics are somehow suspended, only to find that the magic bullet is, in fact, a blank.
Technically, Blockchain includes some neat ideas, even if most of them are decades old. And it’s not difficult to envisage niche situations where the technology could be useful, but operating in modern, high-volume financial markets is not one of them.
Criticisms of Blockchain usually focus on technical issues around whether or not it can ever be scaled up to handle real-world markets, without compromising key principles. The jury is well and truly out on that question.
But here the question is much more basic, can the Blockchain paradigm handle the complexity of modern financial markets?
But first, what is Blockchain?
Blockchain has been called a “distributed ledger”. Think of that as an endless bank statement with each transaction (debit or credit) recorded in sequence from the first to the latest, but without the balance on the right-hand side!
So how do you find out your account balance (or in financial market terms, your position or holdings)?
Well, you have to start at the beginning and add up all of the pluses and minuses to get your latest balance. Since this is clearly time-consuming, what happens in practice is that people keep their latest balances elsewhere, in what are called “wallets” which, importantly, are not part of the Blockchain database model, and hence not covered by the safeguards inherent in that model.
Why are balances not held in Blockchain?
Because a balance changes every time a transaction is made. Balances are mutable, meaning they are liable to change and Blockchain does not support mutability. Thou shalt not change anything stored in Blockchain is the first commandment.
Blockchain deals in individual, immutable and independent transactions and so does not support a fundamental concept of finance called “fungibility”. Fungibility means that every dollar in your account is the same, whether it was from your first pay packet or from this month’s salary. A dollar is a dollar, wherever it comes from. But not in Blockchain!
Say you bought 75 shares of IBM last month and another 75 shares last week, and you wanted to sell 100 shares today. Easy with a stock broker, but tortuous in Blockchain, because you don’t actually have 100 shares, you own one group of 75 and another unrelated 75 shares.
In financial markets, this is where brokers come in. If you have an account with a broker, they know that you have 100 shares (wherever they originally came from) and can find a buyer (or maybe several buyers) who is prepared to buy a total of 100 shares, for a price that is acceptable to you. In some markets, brokers, are prepared to buy at an acceptable price and then find sellers later, because brokers know what is going on in the market. Brokers sort out the real-world problem that individual buyers and sellers do not always have exactly equal and opposite financial needs.
For that service, brokers charge a fee, or quote a “spread”, which pays for finding and matching those buyers and sellers who are actually empowered to buy and sell. In a Blockchain-driven market, without trusted intermediaries, buyers and sellers would individually incur the ‘search costs’ in finding suitable and trustworthy counterparties. In a mature market, these costs are in practice much higher than broker fees.
What Blockchain does NOT do
Blockchain deals with transactions completed in the past and has nothing to contribute to real market concepts, such as price, value, profit or loss and risk.
Taking our example of having bought 100 IBM shares (however they are accumulated). Having 100 shares is pretty irrelevant, unless we know what the shares are worth.
The problem is that, in a heavily traded market, prices and hence the values of assets, such as IBM shares, change millisecond by millisecond. The value of our shareholdings then is mutable so cannot be reflected easily in Blockchain terms unless each change in value is recorded as a separate transaction.
In financial markets, no-one cares what the value of our shares was 10 minutes ago, only what they are worth now - this instant. While a history of price changes is of general interest, and should be recorded, the hypothetical value of our share portfolio at each instant when a price changed in the past is of little note. An immutable history is just not needed.
But while we are interested in value, we are more interested in our profit or loss, or the change in value of our whole portfolio of stocks. For a retail customer, profit or loss is usually calculated on a daily or even monthly basis, but for market players it is calculated in real time, second by second as prices change. Again profit or loss is a volatile, mutable quantity that is not suited to recording in a Blockchain database and again a history of profit or loss on each price change is not needed.
A component of profit and loss that is vitally important is the income that is earned or the expenses paid when holding a position (for example: interest, dividends, bond coupons and option premiums). However, income or expenses are based not on transactions but the positions that were registered on a particular value date, which may or may not be the same as the transaction date.
Other events may happen that change a balance but may or may not be reflected with a transaction. The most obvious of these is a “stock split” or “reverse stock split”, where the sum of all transactions no longer equals the new balance after a split is actioned.
Another example of a non-transactional change is the unexpected redemption of a callable bond by an issuer, extinguishing the position (and by inference all of the transactions related to the position). Other examples include, the exercise of an American option by an investor, which will affect the risk in a portfolio. These examples show that there is not a simple correspondence between a set of transactions and the financial value of the impact of those transactions. In other words, in these instances, the Blockchain on its own does not reflect financial reality.
But it doesn’t end there. While knowing the latest profit or loss is important, it is equally important to know what the risk in the position is. That is, how much profit or loss could be lost in the future, if the position is maintained.
Again, risk is a mutable quantity, which changes not only when prices and values change but also as a result of other factors, such as the range (distribution) of historical price changes, interest rates, currency rates and even qualitative information, such as market commentary.
However, risk is not an asset, it is a concept that only has real value, if it is hedged, or traded using derivatives.
Hedging and Netting
Blockchain deals with individual, unchanging transactions and there is no concept of linking transactions, because that implies mutability when transactions are dynamically linked and unlinked. If one wants to hedge a position (for example reducing or even leveraging risk), one can buy (or sell) hedges, such as exchange-traded futures or options. The portfolio then becomes not a ledger of unrelated transactions but, for trading purposes, a single position, with mutable values, profit or loss and risks (based in turn on the mutable prices, values, profit or loss and risk of the underlying assets).
Linking of transactions is an integral part of modern financial markets. For example, in settling transaction or managing the risks in derivatives positions, it is not the gross value of the underlying transactions that matters, but the net of a particular set of transactions. In mature markets, concepts such as netting by novation (replacing multiple transactions by a new one), and portfolio compression (doing the same for multiple portfolios to reduce systemic risk) have been developed to reduce credit, operational and liquidity risks.
And, mechanisms, such as margining and collateralization, have been developed to minimize credit risks that arise because a counter party may default on their obligations. Of course, margins and collateral values are mutable, changing as external markets change.
Immutability then is not an absolute requirement of financial markets. However, retention of an audit trail of changes to a transaction, and related information, is indeed a requirement and this has been handled by paper and electronic technology since the early days of finance.
Evolution of the Interest Rate Swaps Market
An example of how successful financial markets grow is the evolution of the global Interest Rate Swaps (IRS) market, through various twists, turns and crises, as the swap became the primary mechanism for managing interest rate risk over the past 35 years.
The first IRS contract was a one-off, bilateral contract negotiated in the early 1980s, between IBM and the World Bank. It’s the type of contract that could possibly be recorded in a Blockchain, albeit that the signing was at the endpoint of many months of tortuous negotiations on contractual terms and conditions. IRS is a very successful financial innovation that took off because it reduced overall credit risk, by trading changes in interest rates (an artificial asset) rather than by lending real money.
But before long, the IRS market hit a brick wall, because bilateral negotiations were expensive. As a result, the International Swaps Dealers Association (ISDA) was formed to standardise the terms and conditions of swap contracts – a job that it is still doing well 30 years later, as financial markets continue to innovate.
After several more years of growth, credit risk became a constraint in the IRS market and the concept of a “netting agreement” was developed, effectively treating many IRS transactions as one. In Blockchain terms, transactions were discharged or deleted and replaced by a separate new contract.
Later in the evolution of IRS, the ISDA “credit support annexe” was developed that permitted collateralization to further reduce credit risk. Importantly, this innovation allowed many more parties to become part of the swaps market, which had been constrained to that point by credit risk. The market exploded and according to the Bank for International Settlements, contracts totalling some US$320 trillion in notional value were outstanding in 2015, up from less than US$50 trillion in 1998.
It has been innovations such as netting by novation and collateralization coupled with standardisation of terms and conditions for different types of contract that has enabled this important market to grow. Bilateral contracts have given way to multilateral standards, in which trust is embedded.
In modern financial markets, it’s not the transaction that is important but the net fungible position that is critical. Since Blockchain exists to ensure certainty and immutability, rather than flexibility and change, the concept would have been a deterrent rather than an enabler of growth in such an important market.
Financial markets grow by innovation and, over centuries, new financial tools and techniques have been developed, enhanced and combined to reduce risks and increase trust in markets. Practitioners know that no system is perfect – people are smart – and so trust is forever a concept that must be actively managed.
Those who claim to have found a magic bullet that will ensure trust for ever are naïve. They need to understand how real markets grow and prosper rather than concentrate on the minutiae of recording one of the first, and in most cases the least onerous, step - the initial transaction.
Blockchain is to finance as bricklaying is to architecture. The bricks are of course important, but it is the myriad of ways that the building blocks can be combined into new, exciting and useful structures that is important.
The individual bricks get lost in the overall aesthetic of the building. In finance, transactions are forgotten once they are placed in position, and the efficiency of the overall structure becomes the most important consideration.