- November 3, 2016
- Posted by: IshwarChidambaram, CFA
- Category:BLOG, ExPress
Contributed by: Ishwar Chidambaram, CFA
The advent and spectacular rise of High Frequency Trading (HFT) has been the watershed event, transforming both developed and emerging financial markets beyond all recognition. The jury is still out on whether this represents meaningful progress or is ultimately regressive. But one thing is clear‐ global financial markets will never be the same again.
Gone are the days when traders would pick up a fixed line telephone and whisper in hushed tones “Blue Horseshoe loves Anacott Steel”! Now an algorithm would trade a million shares of Anacott faster than the Gordon Gekko can blink, capturing slivers of profit on thousands of such buy/sell transactions. The HFTs’ need for speed has resulted in flash trades and concomitant flash crashes triggered by rogue algos.
In one instance a rogue algorithm (“algo” for short) consumed fully 10% of entire available communications bandwidth, simply to place humongous numbers of buy and sell orders which were all cancelled, and subsequently vanished without a trace. More destructive are examples of Knight Capital, which lost upto $ 10 MM per minute due to a rogue algo that was to have been deactivated. We have been extremely fortunate that such incidents haven’t occurred in a Systemically Important Financial Institution (SIFI) until now. Regulators have thus far adopted an ostrich‐like mentality, preferring to use “bicycles to catch Ferraris”. However, that needs to change. While not all algos are rogue, it only needs one rotten apple to spoil the entire bunch.
Trillions of dollars have been wiped out of global markets in flash crashes triggered by rogue algos, most notable being the 2010 crash in the Dow and Nasdaq. That episode was blamed on a single trader (Navinder Singh) who used spoofing algorithms to place bearish bets to the tune of $200 MM, replacing or modifying orders 19000 times before they were cancelled. Another example is the Tweet Crash of 2013, wherein a false rumour of an explosion in the White House was translated by algos into a sell signal, leading to a temporary market crash and spikes in volatility.
HFTs need to be regulated in the interest of the greater good, for while they are good servants, they have also proven to be dangerous masters. Exotic and unsavoury elements associated with HFTs, like dark pools and pack hunters etc. only serve to reinforce the need for more and smarter regulation.