Skip to main content

A flash crash is a very rapid, deep and volatile fall in security prices occurring within an extremely short time period.NicoElNino/Getty Images/iStockphoto

You have likely heard the phrase "flash crash" in recent years.

Last Thursday night (Friday morning in Tokyo), we saw a flash crash in the British pound. A flash crash is generally defined as a very rapid, deep and volatile fall in security prices occurring within an extremely short time period. A flash crash frequently stems from trades executed by "black-box trading," combined with high-frequency trading, whose speed and interconnectedness can result in the loss and recovery of billions of dollars in a matter of minutes and seconds.

Today, computers are responsible for the vast majority of trades entered into the marketplace. On the equity side, gone are the days of the specialist who was responsible for keeping orderly markets. The job made sense when the spread between the bid versus the ask was wide enough that the market maker had an economic profit to take the other side of the trade. The primary reason we have seen an increase in flash crashes is that there is no longer an economic profit to stand in front of a huge market order.

The black-box and high-frequency trading styles are driven by computer algorithms (algos) that most investors ultimately use to route orders to the exchanges, and are coded to make money for their users. When volatility spikes for any reason (geopolitical risk, headline news, earnings updates, unpredictable event known as "black swan," "fat-finger" trades, etc.), these algos are smart enough to widen out their bids and asks or they disappear altogether. The markets then hit these periods of illiquidity when the algos that are still active seek the next bid or offer to buy or sell – often triggering stop loss orders. In the foreign exchange markets, there are often billions worth of stop orders sitting in the book at any time. Stop loss orders typically become market orders when hit – thus further exaggerating the illiquidity.

French President François Hollande started speaking last Thursday night in Paris – just after 8 a.m. (Friday) Tokyo time (or 7 p.m. Thursday ET). One remark he made was Britain would "pay the price" for choosing Brexit. Some had attributed the reaction to a fat-finger trade. A fat-finger trade can be viewed as a trader wanting to sell, say, £100-million ($146-million) but inadvertently enters the order as £100-billion. There tends to be a brief period of low liquidity as Wall Street goes home and Tokyo takes over the book around that time, so the illiquidity at the time was even more acute. About the only markets that are active at that time are the foreign exchange markets, so that is where all the computers tend to be focused until equity markets open.

We have seen several prominent flash crashes in recent years:

  • May 6, 2010, was the most famous and topic of the book Flash Boys by Michael Lewis;
  • April 23, 2013, when The Associated Press’s Twitter account was hacked and hoax tweets about the White House being attacked hit the headlines;
  • Aug. 24, 2015: We saw the Dow Jones industrial average open down 1,000 points as anxiety over Chinese currency devaluation caused a period of illiquidity.

There are many less famous examples where so-called black swan headlines hit the markets and many of these computerized algos go haywire.

For individual investors, expect that these types of events will continue to happen with increasing frequency and that there is little you can do to forecast when they might happen. The best way to protect yourself is not to use stop loss orders. A stop loss order is typically placed on a stock or an ETF that you own at a point that you would not want to own it to limit your losses.

This applies for entering market buy orders, too. In 2009, a series of unfortunate events caused a computer to enter orders on ETFs 20-per-cent to 25-per-cent higher than current market levels. The exchange closed trading and when the market re-opened, many investors ended up with very bad trades. If you like to use stop loss orders, be sure that you use stop limit orders. This is where you set a limit on what price the order would be exercised if it was tripped. That would avoid getting a really horrible fill and still limiting losses.

If you want to buy, always set a limit price maybe a few pennies higher than the current bid, so in case of a moment of algos going off the rails, you don't pay up for something at a ridiculous price.

Larry Berman is co-founder of ETF Capital Management. He is a Chartered Market Technician, a Chartered Financial Analyst charterholder and is a U.S.-registered Commodity Trading Advisor.

Interact with The Globe