On the surface, bid-ask spreads are far more narrow than even several years ago.

However, during periods of financial stress liquidity seems to vanish. Also, the increasing amount of fragmentation (i.e. new exchanges, crossing markets / dark pools, etc.) is moving markets away from deep central pools of liquidity. Combined with ultra-HFT linkages across markets there seems to be a potential for increased systemic risk via contagion. There also seem to be glitches -- such as today's 9% flash crash in Apple based on a 100-share trade. There's also research that at the micro-structure level price changes are now exhibiting non-normality (where previously this was observed only at lower frequencies).

The evidence seems mixed and anecdotal. How does this net out? Is there are any hard research on the effects of HFT on market depth, liquidity, and volatility during peacetime and periods of stress?

  • 1
    $\begingroup$ Tal asked a question about HFT last year that generated a list of articles: Has high frequency trading (HFT) been a net benefit or cost to society? $\endgroup$ – chrisaycock Mar 23 '12 at 19:06
  • $\begingroup$ Yep - I looked at that. That asks net benefits to society. This is more specific as to effects on the market but thanks for the pointer. $\endgroup$ – Ram Ahluwalia Mar 23 '12 at 19:29
  • $\begingroup$ HFT helps speed up the convergency process. $\endgroup$ – user2204 Mar 24 '12 at 4:13

Joel Hasbrouck (imho, a leading expert in market microstructure) has a paper on this:


From the abstract:

Our conclusion is that increased low-latency activity improves traditional market quality measures such as short-term volatility, spreads, and displayed depth in the limit order book.

  • 1
    $\begingroup$ Thanks! Funny how all these papers seem to be contradict each other. $\endgroup$ – Ram Ahluwalia Mar 23 '12 at 21:31
  • $\begingroup$ It does not exactly contradict the other papers, it all depends of the measure used and if you test for liquidity during stress periods. $\endgroup$ – BlueTrin Nov 2 '12 at 14:14

A race to zero by Andrew Haldane, Bank of England has some interesting content and references to research regarding this.

  • $\begingroup$ Interesting paper and well written. Seems to contradict the Joel Hasbrouck paper. $\endgroup$ – Ram Ahluwalia Mar 23 '12 at 21:32

I have been coming across a few of these research lately.

I don't have access to it anymore but I remember reading the abstract of this article, but it's really outdated now.

After a quick look, I found this paper which is much more recent (2011) as well as this one (2010).

  • $\begingroup$ What do these papers say? Can you summarize them here? $\endgroup$ – chrisaycock Mar 23 '12 at 19:04
  • $\begingroup$ AS I said in the post, I can't access them anymore... I'll try to see if I've got something in my notes... $\endgroup$ – SRKX Mar 24 '12 at 1:06

The problem with most academic studies is that they do a poor job of defining what is "HFT", deterministically attributing a structural change to "HFT" and removing the effects of extraneous factors (other algorithmic trading activity such as SORs, large trading by discretionary participants).

One of the best modern studies of the market structural effect of "HFT" owes to Tokyo Stock Exchange's introduction of Arrowhead in January 2010, which effectively enabled "HFT" in the venue. This is a particularly useful data point because this was carried out by a modern exchange with an earnest effort to monitor the effect of the change and "HFT" participants in a recent date.

Market depth and liquidity: It was found that the introduction of Arrowhead co-occurred with an instant increase in depth of orders at the best bid and offer of between 62% to 139% in all symbols.

Volatility and liquidity: TSE also found that they were now able to distinguish between colocation and non-colocation participants and analyze their flow during periods of high volatility. It was found that during market downturns, non-colocation participants were net sellers whereas the colocation participants were net neutral. As a result, the volatility has to be exacerbated by non-HFT participants and we can infer that non-HFT participants are net liquidity takers during periods of volatility while HFT participants are net liquidity providers during periods of volatility.



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.