On the morning of the January 30 at 11.20am, trading in HSBC shares on the London Stock Exchange (LSE) was halted after the share price spiked up by 10% in the space of 25 seconds. There was no news or announcements which could have affected the share price, and when the stock re-opened after a five minute auction, it carried on trading around the same levels as before the spike, ending the day up slightly at 630.2p. So what happened?
By Rupert Osborne who works on IG’s trading floor in London. Learn about CFD trading.
On the morning of the January 30 at 11.20am, trading in HSBC shares on the London Stock Exchange (LSE) was halted after the share price spiked up by 10% in the space of 25 seconds. There was no news or announcements which could have affected the share price, and when the stock re-opened after a five minute auction, it carried on trading around the same levels as before the spike, ending the day up slightly at 630.2p. So what happened?
The last ten years have seen an explosion in the proportion of trading done by automated processes, often termed algorithmic or high-frequency trading, as traders use systems with a variety of programmed and real-time inputs to automatically place orders into financial markets. They may attempt to profit from small movements in prices multiplied over thousands of trades, or hide a large order and try to get the best price by working it in small pieces.
It is impossible to know the full scale of automated trading, but by some estimates, it accounts for more than half of all orders placed on the LSE. It has been under increased scrutiny in recent years, especially since the flash crash of May 6, 2010, when financial markets in New York were rocked by massive unexplained declines, which were quickly reversed. The benefits that automated trading brings to trade execution are countered by the increased risk that they can de-stabilise markets, with potentially serious consequences.
The spike in HSBC’s share price on January 30 appears to be an automated trading system which has gone dramatically wrong, and using trade data I have been able to plot the two graphs below, which zoom into the 25 second period when the spike happened. In the first chart, you can see that between 11.20:17 and 11.20:22, the price increases from 629.2p to 635p, as 600,000 shares are traded. The price action then becomes increasingly erratic, as the automated system continues to place buy orders, and the volume traded increases steadily. It hits a high of 688p at 11.20:42, before the LSE’s systems halt trading and the stock goes into a five minute ‘volatility auction’, which is meant to allow order to return to the market after a stock has moved 10% from the previous day. After the auction, HSBC re-opens at 630p.
During this period, 1,890,660 shares were traded at an average price of 648.224p. HSBC’s daily volume over the previous 12 months averages 21.05 million shares, meaning that over 9% of a normal day’s trading happened in just 25 seconds. Based on the re-opening price of 630p, whoever was responsible for the huge volume of buying was sitting on an immediate loss of £344,562.
By looking into the spread of the best bid and offer in HSBC’s shares during the spike, you can see how aggressively this order moves the market. Normally the spread between the two is around 0.1 – 0.2p, and during the initial movement the spread is fairly steady. But as the huge volume of buying continues, the spread widens to as much as 10p. You can see that the best offer price keeps spiking up and away from the best bid, meaning that the buyer is continually lifting the shares on offer until the market is halted. Interestingly, there are some large spikes down in the best bid, especially at 11.20:40; presumably someone else in the market realised what was happening and used the opportunity to sell some HSBC stock at an extremely advantageous price.
After the stock re-opened, there was an investigation by the LSE, but the exchange did not cancel any trades. Based on the data above, it seems unlikely that a human trader would have executed this kind of huge order (known as a ‘fat-finger’ error), as there would have been ample time for them to cancel it. Instead, an automated trading system seems to be the culprit, and continued buying stock until the LSE halted trading. During the spike, other markets were also affected, as other traders and automated systems noticed the movement and traded on the basis of it. HSBC makes up 7% of the FTSE 100, and during the spike FTSE futures moved up by around 30 points as below:
While it seems improbable that automated trading will disappear, bearing in mind the technological advantages that it can bring, any kind of trading system that can move the valuation of the UK’s largest bank by over £11 billion in 25 seconds should be treated with caution. Whoever was responsible for this spike, and suffered that enormous loss on their trades, has certainly learned that lesson.
Spread bets and CFDs are leveraged products. Spread betting and CFD trading may not be suitable for everyone and can result in losses that exceed your initial deposit, so please ensure that you fully understand the risks involved.
This information has been prepared by IG, a trading name of IG Markets Limited. The material on this page does not contain a record of our trading prices, or an offer of, or solicitation for, a transaction in any financial instrument. IG accepts no responsibility for any use that may be made of these comments and for any consequences that result. No representation or warranty is given as to the accuracy or completeness of this information. Consequently any person acting on it does so entirely at their own risk. Any research provided does not have regard to the specific investment objectives, financial situation and needs of any specific person who may receive it. It has not been prepared in accordance with legal requirements designed to promote the independence of investment research and as such is considered to be a marketing communication. Although we are not specifically constrained from dealing ahead of our recommendations we do not seek to take advantage of them before they are provided to our clients.
