While we at OmiCronFX have spent much time in developing our Forex
trading software so that it can be consistently profitable, we know that it is
of vital importance that we also concentrate on the risks that are becoming
more and more apparent in relation to trading in general, and algorithmic
trading in particular.
Computers have the ability to act and
react rapidly – that is the point in developing and using algorithmic routines.
Unfortunately, this can be double edged sword, which will be obvious in a
situation where losses, instead of profits, are in the course of being rapidly
multiplied, for whatever reason. So operators of trading software have a real responsibility
and incentive to thoroughly test their creations, and to implement the
safeguards that have been built up over the years to protect systems where
software is being put in place anywhere, not just for Forex or stock trading. The
faulty implementation of accounting software in a manufacturing company, for
example, can be just as disastrous for that company, in terms of its ability to
safely manage its business, as the out-of-control buying of a losing security
by a hedge fund when its High Frequency Trading software malfunctions.
And, as far as trading is concerned, computers
are not just in play in the HFT trading funds. They also now control and manage
everything that happens at the brokerage, exchange, market-making, and price and
exchange-rate determination level, and even for compliance with trading
regulations by the SEC and similar bodies around the world. These computers also
have the capacity to malfunction, as has been shown in the case of the so
called “flash crash” of May 6th 2010, when the Dow Jones Industrial
Average dropped over 1000 points, or 9%, in a matter of minutes. One factor in
the events leading up to the crash were so-called “technical glitches” in the
reporting of prices on the NYSE that tended to lead to delays in quotes, with
subsequent wholesale confusion for traders who mistakenly thought they were in loss
making situations, which they naturally attempted to liquidate.
Ignoring
algo is not an option
The cumulative effect of all of these
interlinked computers for the operation of the markets can lead to the extraordinary
amplification of what would otherwise be notable, but not necessarily disastrous,
occurrences. The best example of this in
recent times was the melt-down in the exchange rate between the Euro and the
Swiss franc on the occasion of the sudden removal, by the Swiss Central Bank,
of the cap it had maintained on the value of its currency to prevent its rate
against the Euro (shown by the EURCHF currency pair) going below 1.20 (a fall
in EURCHF means a strengthening Swiss franc [CHF]). Many Forex players had been
so convinced by the strong rhetoric of the SNB prior to this, to the effect
that it would defend the cap by whatever action was needed to do so, that they
had used high leverage to take long positions in EURCHF (betting it would go
up) as it approached the level of the cap at 1.20. Some of these funds were
wiped out when the SNB announced that the cap was being abandoned. A number of Forex
brokers were placed in receivership, while others had to absorb losses running
to hundreds of millions.
The lesson from all of this is that
computers and algorithms are already ubiquitous in trading. The dwindling
number of manual traders will find two things: (1) they are at a disadvantage because
of the sheer speed, lack of emotion and infinite patience of their algorithmic opponents’
computers and (2) they are at the mercy of possible malfunction of the computers
that they must interact with in any event, and / or the magnification of effects
that these machines can bring about.
No comments:
Post a Comment