You all remember Andrei Kirilenko, MIT professor and former CFTC Economist? You should, as he was at the forefront of the SEC/CFTC’s Flash Crash analysis, as well as the author of numerous HFT-related academic studies, including one that the CME forced the CFTC to silence. Most of his studies deal with the study of volatility, and especially the effect HFT has on it. From the SEC/CFTC Flash Crash findings:
“Indeed, even in the absence of extraordinary market events, limit order books can quickly empty and prices can crash simply due to the speed and numbers of orders flowing into the market and due to the ability to instantly cancel orders. Liquidity in a high-speed world is not a given.”
Kirilenko , Richard Sowers, and Xiangqian Meng have together just published a new study on high frequency trading in Algorithmic Finance, titled A Multiscale Model of High Frequency Trading. This is an interesting study because it attempts to show the different types of effects that two different types of traders have on limit order books – and specifically volatility. Inside of the equity trading industry, so many of us are told that HFT does not affect slower fundamental traders:
“They go at the speed they want, and if you don’t need that speed or use the same strategies, then you need not concern yourself with the order types and faster speeds. In fact you are benefiting because their activities are giving you more liquidity and tighter bid ask spreads.”
This paper challenges that, and does so in a macro way. It starts with the assumption of the multi-scale model – that limit order books are occupied by low frequency fundamental traders and high frequency traders who operate at different speeds. It ends with an interesting conclusion – that HFT adds to stock volatility daily – and not detracts from it – which contradicts the studies sponsored by HFT firms and Stock Exchanges.
This study demonstrates that even assuming the “uninformed” nature of HFT traders on exchanges, their interaction with low frequency traders specifically has a quantifiable effect on stock volatility:
Our analysis suggests that there should be something like an ‘HFT-premium’ in volatility for stocks which are traded by HFT participants.
Practical implications? Perhaps Piper Jaffray’s Gene Munster and Topeka Capital’s Brian White need to retool their analysis of Apple common stock. You know – to take into account the HFT Volatility Premium.
The SEC’s Gregg Berman often cries for the need for real studies at the SEC, as opposed to claims from pundits and fear-inciting media. We hope he personally examines this study.