The collapse of Archegos shows how important it is to understand data

Extreme liquidity, combined with high levels of market volatility and high-tech valuations, can make a lethal cocktail. The collapse of Archegos Capital is proof of that. It was this combination of extreme liquidity, high volatility and high valuations that ultimately led to the margin call collapse and the downfall of the firm

Part of the blame can undoubtedly be put at the door of delays in implementing the Dodd-Frank Act in the US. Those rules — designed to shore up big banks and manage excessive risk-taking in the derivatives market — had been repeatedly put off. If they had been implemented more rapidly, they could well have helped to manage this crisis.

However, it isn’t the underlying issue. Although this may have been a case of excessive risk, it is important to recognise that in the financial services business, we will always need to take risks — finance is in the business of taking measured risk and generating returns.

READ Credit Suisse’s main unit downgraded by Moody’s on Archegos, Greensill fallout

The situation we’re left with is that banks are forced to continue to take risks, despite an incredibly challenging market with low yields but growing and stringent regulation. To manage that risk, banks absolutely must be able to understand the data and the physical documents underlying their positions.

Of course, some banks have systems in place to stop risky trades before they cause problems. But others do not. The have-nots aren’t on top of monitoring disclosure and putting in place automatic risk management procedures at all.

We recently spoke to one bank that, once we had an opportunity to analyse some of its data, realised it actually held positions it wasn’t even aware of. This activity is taking place out of sight, which could be very damaging.

That lack of understanding is incredibly concerning, given the huge impact the Archegos disaster ended up having on major global banks.

Major banks have tens of trillions of dollars’ worth of exposure to the derivatives market at any one time. Of course, most of that is well-hedged and managed. However, all those positions are underpinned by similar agreements and so if there are any small changes made, they could have huge implications for the overall exposure of that bank.

Fundamentally, if banks don’t address this issue, something like Archegos will happen again — and it won’t take long.

To manage that risk then, they must get an understanding of their data. The difficulty is that there’s a reason the banks haven’t got a handle on the data so far: 98% of the world’s data has never been analysed and, more strikingly, 90% of the world’s data is unstructured, according to McKinsey. That means it’s almost unusable.

If you’re an incumbent bank, that figure is likely to be even higher. Unlike fintech businesses or big tech firms, you won’t be a digitally native business with digital infrastructure from day one.

Banks need to use artificial intelligence to access that data and then again to act on it. Of course, it could technically be done manually; however, the cost and time implications of that are huge.

We have recently been helping a number of banks with transition away from the London Interbank Offered Rate. With the number of contracts that need to be analysed, it would cost hundreds of millions of dollars in labour to do that manually; it can be done in days with AI.

READ Regulators tell banks it is time to start using SOFR and ditch Libor

That speed isn’t just important for cost, but also in the case of scenarios such as Archegos. Obviously, the ideal scenario would have been if the banks knew where they were exposed in advance and were therefore able to prevent it from happening entirely. However, if something had slipped through the net, AI would have enabled them to react instantly. Running all the necessary contracts through an AI platform to draw out the relevant data would have meant a much quicker reaction and inevitably a much less painful outcome.

I started my career in financial services at the height of the financial crisis in 2008. Coming from a scientific background, it became obvious to me that data is hugely underused in the sector. There is a great scene in the film The Big Short, where Michael Burry is manually going through hundreds of massive loan and debt documents. That eventually enables him to make the determination to short the market. AI technology now exists that means it is possible to do all that automatically — the data that is hidden in thousands of contracts and loan agreements can be drawn out into a usable format.

The Archegos disaster revealed a glaring problem in the way institutions look at risk management. But the solution is there. Any banks that aren’t using technology to help in this are choosing to do so in a way that verges on negligence. The solution is AI — it is no longer a “nice to have” in banking, it’s a must-have. Until that is understood, the problem will only get worse.

Lewis Liu is co-founder and CEO of Eigen Technologies.

Most Related Links :
editorpen Governmental News Finance News

Source link

Back to top button