Algogeddon: Coming soon to an FX market near you!
We’ve had some famous examples over the years, from Herstatt in 1974 through to LTCM in 1998, where one position in the market got so big that it’s liquidation caused massive disruption and huge spikes in volatility.
We live in different times now but the risks are still the same. The present day market is dominated by huge ‘Algos’, systematic models which thrive in range-bound markets where the models can pick up the spread. Even if the spread is very tight, if you are picking it up thousands of times a day in significant volumes, you are certain to make a lot of money.
Old timers will remember the markets of the 1980’s and 1990’s when the bigger players could bully the market (most of the time); they could soak up whatever was being thrown at it and then simply move the market because they were big enough to do so. But these were only ever sort-term movements as no one player had the credit lines and depth of pockets to maintain dominance for any length of time.
Now we have a situation where the Algos are so huge that they can in essence do exactly what the bully boys did 30 years ago. They can tie the market into a tight range, warehousing the huge underlying positions which are being generated, and keep generating spread revenue session after session.
But, (there’s always a but) there is a problem. All the Algos are programmed the same way and do the same thing. Banks and hedge funds simply turn on the Black Box and let the profits flow in. Remember how this ended for LTCM! The main impact in the case of LTCM was in USD/JPY which fell something like 15 big figures in an afternoon! Now extrapolate this scenario to where you have multiple LTCM-lookalikes all doing the same thing in bigger size and in less liquid currency pairs. How do you think this is going to end??
I don’t know when it will happen but I do know that it will. Take it away Shirley, it’s all just a little bit of history repeating.