Flash Crashes – time to stop knee-jerk blaming of HFT?

24 May 2017


On the 6th May 2010, Wall Street experienced what quickly became known as the ‘Flash Crash’, in which markets, including the S&P 500, the Dow Jones Industrial Average and the Nasdaq Composite, collapsed and rebounded almost instantly.  On the 15th January 2015, the Swiss Franc experienced a similar event against the Euro. And in October 2016, there was a flash crash in sterling, following Britain’s vote to leave the EU.

The recent events in the sterling market seemed like a good idea to look back at the post event analysis of previous ‘flash crashes’ and review the findings.

If you were to browse your Twitter feed or read newspapers in the immediate aftermath of each event, you’d notice fingers being pointed firmly at High Frequency Trading (HFT). In the rear view mirror of hindsight, and thanks to a number of regulatory bodies undertaking detailed analysis of each event, we now know these claims to be knee-jerk reactions, made with a surprising lack of supporting evidence. So what has caused these flash crash events? Let’s step back and take a look at the facts.

During the Wall Street Flash Crash in 2010, several indices experienced an unprecedented single day fall. The Dow Jones Industrial Average Index fell by six percent in a matter of minutes and "investors saw nearly $1 trillion of value erased from U.S. stocks in just minutes.” In the immediate aftermath many were quick to blame high frequency traders for accelerating price moves. However, extensive research into the causes by U.S. regulatory bodies the SEC and CFTC exonerated HFT from causing the crash (although lessons were identified such as the benefit of ‘circuit breakers’ to help traders ‘coordinate their liquidity supply responses’).

And following an extensive investigation by US federal prosecutors, UK based trader Navinder Singh Sarao (who traded from his parents’ home) was extradited to the U.S. where he was accused of creating an “extreme order book imbalance” which allegedly affected stock markets and exacerbated the flash crash.

In October 2016, sterling suffered a ‘flash event’ and once again there was speculation that HFT was to blame.  It was alleged that an algorithm set to scan the news for negative Brexit stories went into overdrive. The reality is quite different. A report by BIS found ‘a confluence of factors catalysing the move, rather than to a single clear driver.’ The report also found that the time of day was a significant factor in increasing the vulnerability of the sterling foreign exchange market. Further, it noted that events such as this have been short lived and have not significantly impacted financial stability. 

Bank of England paper found that the Swiss National Bank’s (SNB) decision to no longer hold the Swiss franc at a fixed exchange rate with the Euro led to a flash crash in 2015. It was the SNB’s surprising decision to remove the cap that initiated the ensuing chaos, and not HFT.

We often talk about volatility in financial markets, but we must recognise that we now live in a volatile political climate. Decisions taken by voters in 2016, in Britain and the U.S. in particular, had a significant impact on markets across the globe. With these results in mind, and with UK and German elections to come this year, it would be naïve not to expect further market volatility and price gaps.  

There are certainly lessons that need to be learned here. HFT firms are one of many market participants and the reality is that algorithmic trading is here to stay, with automated trading technology likely to develop even further over the coming years. To blame firms using HFT for these flash crashes, without closely analysing the evidence, is irresponsible. It’s time to stop jumping to blame HFT and take more time to consider the facts.

  • EPTA
  • Audio
  • Blog