At the risk of giving away the punchline, we don’t think that quantum computing is going to revolutionise financial services just yet — though its impact on encryption is worth being across. Nonetheless, we think there’s value in understanding the basics of the technology, because it’s only going to be spoken about more.
What on earth is quantum computing?
In classical computing — the kind we all know and love — information is stored as bits: 1s and 0s. In quantum computing the bit is replaced with the qubit (or quantum bit) which can take the value 1, 0, or some combination of 1 and 0. So far, so simple. But to explain exactly what we mean by “some combination of 1 and 0”, and how qubits interact with each other, we need two concepts from quantum mechanics:
1. Superposition. As made famous by Schrodginer’s simultaneously dead and living cat, at quantum scales particles can exist in multiple states simultaneously, only resolving into a single state when measured. Qubits exhibit this property, existing as a superposition of 1 and 0 until a calculation is complete, at which point they definitively become 1 or 0.
2. Entanglement. Described by Einstein as “spooky action at a distance”, entanglement is when the state of every particle in a group becomes dependent on the state of every other particle. Entangling qubits in this way is likely key to achieving fast quantum computations.
There is a class of mathematical problems that currently take classical computers an inordinate amount of time to solve — often billions, if not trillions or quadrillions of years. Take encryption. RSA (Rivest–Shamir–Adleman) is a cryptographic system still widely used for the encryption of web traffic and financial transactions, and can be broken by solving mathematical problems involving the factorisation of prime numbers. Cracking RSA-2048, a commonly used version, with a classical computer would take far longer than the current age of the universe. In theory, if we could build a perfectly stable quantum computer approximately 4000 qubits, we could get through RSA-2048 in something closer to 10 seconds.
For now, though, our WhatsApp messages and online banking are safe (from quantum computing, anyway). Today’s quantum computers are nowhere near being able to crack RSA-2048. The biggest quantum computer at the time of writing had 1180 qubits, which doesn’t seem a long way from 4000. The difficulty comes from the requirement for perfectly stable qubits in the example given above. Quantum computers are extremely susceptible to noise and error. Unintended entanglement with things other than other qubits leads to error rates in the range of 0.1%-1%. At 0.1%, our 10 second calculation now takes 8 hours and requires something like 20 million qubits. Needless to say, there’s a long way to go.
Quantum computing in financial services?
Nevertheless, quantum computing has made its way onto the financial services conference stage in the past year or so (where it caught our attention). Use cases being discussed seem to fall into two categories:
1. Cryptography and data security. Despite the non-existence of quantum computers powerful enough to crack commercial encryption, some financial institutions are already taking steps to mitigate against “harvest now, decrypt later” attacks. The idea is to move to post-quantum cryptography (PQC) — that is, encryption methods that can withstand attacks from quantum computers — now, lest attackers harvest data and decrypt it when sufficiently capable quantum computers become available. PQC already exists, with Apple introducing it to iMessage in early 2024, and HSBC announced a trial of PQC in 2023.
2. Market simulation and portfolio optimisation. Evangelists claim that quantum computing may be able to speed up market simulation and portfolio optimisation, allowing calculations to be run more frequently and with a larger number of input parameters. IBM suggests that quantum computing may assist investment managers to “incorporate real-life constraints, such as market volatility and customer life-event changes, into portfolio optimization.”
Moving to PQC seems like a logical thing to do. Indeed, the Australian Signals Directorate (ASD) has been encouraging organisations to “create a transition plan for the use of PQC algorithms within their environment, including the testing and adoption of new PQC algorithms as well as the decommissioning of legacy cryptographic algorithms” since at least 2022.
Talk of using quantum computers to run market simulations and optimise portfolios seem to us more like buzz, at least for the time being. When useful quantum computing becomes a reality, it will almost certainly be extremely expensive. Most existing quantum computers involve suspending individual ions in a vacuum using electromagnetic fields and lasers, or cooling superconductors to nearly absolute zero (-273.15 °C) — both of which cost a pretty penny. Thus, quantum computing makes a lot of sense for use cases that cannot be realised any other way — like breaking encryption (if you happen to be a cybercriminal) or estimating the ground state energy of a quantum system (if you happen to be a massive nerd). Market simulation and portfolio optimisation can currently be done in a range of different ways, from the deterministic to stochastic, using existing technology. Granted using existing machine learning and AI tools to do this can be expensive, but quantum computing is unlikely to offer a cheaper way to do it.
What’s next?
Quantum computing is still in its infancy. The current approaches to building a quantum computer with their associated limitations could be rendered irrelevant by a major unforeseen development in the field. A “ChatGPT moment” may be just around the corner, and in the meantime, developing a plan to adopt PQC is a good step for organisations to take. Beyond that, we suspect it will be some time before the pure theory of quantum mechanics is sullied by the existence of a fully functional quantum computer — let alone applications of quantum computing to financial services.
This article was produced as part of The Quarterly – Q2 FY25
For more information about anything you’ve read here, or if you have a more general inquiry, please contact us.
Key Contributors:
Kevin Fernandez is General Manager, Market Strategy and Propositions at Novigi, and is based in the Melbourne office.
Sophie Coianiz is an analyst in the Market Strategy and Propositions team at Novigi, and is based in the Sydney office.