16 September 2014

No link between HFT and market manipulation - research

13 November 2012  |  7377 views  |  12 Stock exchange numbers

There is no link between high-frequency trading (HFT) and market manipulation, according to research from Australia's Capital Markets Cooperative Research Centre (CMCRC).

Regulators on both sides of the Atlantic have had their sights trained on HFT ever since the 6 May 2010 flash crash. Last month European parliamentarians voted in favour of a crackdown on the practice as part of MiFID II.

However, the CMCRC research, based on five years' worth of data - from 2006 to 2011 - from the London Stock Exchange and Euronext Paris, has found that the much-maligned HFT does not correlate with an increase in market abuse.

During the five year period covered by the report, there was a considerable increase in the HFT proxy on the LSE and Euronext Paris markets.

There was also a statistically significant negative correlation between the HFT proxy and end-of-day price dislocation alerts - which is a proxy for market manipulation - for both markets, suggesting more high-frequency trading actually equals less market abuse.

The report initially found a correlation between HFT and 'ticking' - incidences of one-share executions moving prices - but this disappeared when the data was controlled for variations in volume and volatility. This may suggest, says the CMCRC, that some of the ticking being blamed on HFT is in fact just a feature of normal market activity.

Professor Alex Frino, CEO, CMCRC, says that despite intense interest in HFT and algo trading since the flash crash, there still remains a lack of empirical research directly examining their effect on market quality and integrity.

"In an environment of such low returns, everyone is going to be looking for a scapegoat. However HFT and its relationship to market fabric is very complex, and needs to be analysed as such before any conclusions can be drawn. It's not good enough just to have an opinion, when regulations are being drawn up that will affect the way markets work around the world," says Frino.

The CMCRC report on market manipulation was commissioned by the Bank of England for the Foresight project on HFT. The final Foresight working paper, which brought together 35 academics from nine countries, found fault with several planks of the EC's planned regulatory changes.

Comments: (12)

A Finextra member | 13 November, 2012, 09:09

I have to agree with the conclusions so far on HFT.I do not believe that HFT is a mechanism for market abuse just a technology that brings unintended consequences. The question is not to ban it but understand HFT and how it can be incorporated within the market without causing distress in other parts of the market. The search goes on

Neil Crammond - evoi - london | 13 November, 2012, 09:13

sorry ; there are very obvious links with HFT  and manipulation ! With 30 years experience its very easy to see abuse and where it comes from and if an industry is claiming 70% share of market it becomes even easier . 

I would accept CMCRC article if they accepted  20% of market abuse but to state none is a fairy tale . My argument is over the last 5 years its become more obvious and even now HFT beginning to admit some abuse ! 

A Finextra member | 13 November, 2012, 09:25

Abuse indicates some deliberate act with the intention of manipulating the market. I do not think that HFT is deliberatly manipulating the market. Unintended consequences are the problem

Steve Leegood - Bryok Systems Limited - London | 13 November, 2012, 11:34

Like CMCRC's study into the cost of insider trading (drawn on by the EC to derive its €13Bn estimate of the annual cost of abuse of EU equity markets) which was arguably fundamentally flawed by the measure it took as a "proxy", so it seems is this. Why are "end-of-day price dislocation alerts" an adequate proxy for market manipulation, and particularly for HFT abuse, when at end of day many HFTers are trying to close out their positions, not manipulate?

I do agree that there "remains a lack of empirical research directly examining [HFT's] effect on market quality and integrity". However, that somewhat misses the point.

IOSCO defines market integrity as "the extent to which a market operates in a manner that is, and is perceived to be, fair and orderly" (my italics). The effort and emotion invested in debating the effect of HFT on market integrity indicates, in IOSCO's terms, that there is a widespread perception - justified or not - that HFT may cause markets to be unfair or disorderly, and hence confidence and participation in the markets is damaged. If perception is reality, and market integrity depends on perception - as well as objective fact - then HFT has a perception problem, and so do markets that support it.

Retired Member - - | 13 November, 2012, 11:34

Depends on your definition of 'abusive'. I would argue that the whole purpose of HFT is to gain advantage by making decisions quicker than other players - what you have created is two streams of trading. 'Traditional' human-based trading and computerised HFT trading. More and more of the human trading is being pushed to OTC transactions (partially for fear of predatory algo's) whilst HFT is dominating the Exchanges.

When one side of a trade profits, there is always someone, somewhere, that loses out. In order for HFT to be profitable it has to take advantage of slower players, which at the very least is an unfair advantage (if not abusive). And this statement assumes all HFT activity is good natured and not intentionally preditorial/abusive......

A Finextra member | 13 November, 2012, 12:51

There is a real problem with HFT that it is must that a solution be found. We operating a two speed market. To a much slower degree than today the market has always been about speed and the capability to see a deal and profit and act before the prices change. This is a fundemental however today we have a structure where speed and technical capability out weighs investing. We have far too much trading that does impact the investor and sometimes adversly. A solution must be found and i do not like banning as a solution. We need a mechanism to better monitor and regulate the market to ensure its fair to all. Maybe some form of HFT tax might limit the use and profitability? What is certain is that the markets can not operate as they are today without defeating the main purpose of why they exist at all

Retired Member - - | 13 November, 2012, 13:56

Here's where I think that the research goes a little awry:

"end-of-day price dislocation alerts - which is a proxy for market manipulation".

Using an 'End-of-day price dislocation alert' as a starting point for HFT manifested manipulation is a non-starter.

HFT stuff happens very quickly.  When the trading bot are gaming the system, they are looking for tiny little, very quick price moves.  

Thus, to research HFT manipulation, I think that you would have to have a significantly more granular definition of market manipulation.

Steve Leegood - Bryok Systems Limited - London | 13 November, 2012, 15:41

Gary suggests "we need a mechanism to better monitor and regulate the market to ensure its fair to all."

We do indeed, and it's currently the panoply of legislation, regulation, surveillance and compliance. Which together cannot agree the nature, size and shape of the problem, let alone define how to address it, or monitor that some ideal new system works properly.

An alternative is to accept what most traders admit, behind their hands as it were, that of course markets are gamed, people cheat, but it's a game for grown-ups. Well so's the drugs trade, but that does not make it acceptable.

If we do accept that, and reluctantly conclude that legislators et al will not sort it before all our pension funds have been cheated out of many many billions, then perhaps we should expect the industry to sort out its own mess.

One immediate possibility is for perceived market integrity - or the lack of it - to be taken into account in best execution.



Neil Crammond - evoi - london | 13 November, 2012, 15:46

steve has a point that we cannot go forward until we can admit the current abuse as real  ! My gripe is that the exchanges have constantly denied "any abuse ! " The cure is there BUT do we want the medicine ?


Bruce Bland - Fidessa - London | 14 November, 2012, 09:04

Having just read the report I have to agree with Carl above. The analysis done takes only suspected cases of EOD manipulation (as found by SMARTES) using presumably statistical methods, and then goes on to use this as a proxy for saying that HFT trading is a good thing! Anybody else see the problem here?

Toby Corballis - Rapid Addition - London | 16 November, 2012, 14:56

I think the title would read more accurately as 'No link found...' and then I might even add 'No link found yet...'. The thing with correlation is understanding how it works. Only then can you say it's there or not. I'm not convinced that the report's authors have got that far. I'm not questioning the ability of the authors per se, but HFT spawns vast amounts of data and it's no trivial matter to work out how to analyse it. It does seem to that they've looked at a very small subset of the data.

Neil Crammond - evoi - london | 16 November, 2012, 15:02

Until we accept manipulation HFT cannot advance . Sadly there is market abuse in all markets and always will have a percentage  as money is involved ! HFT  has taken that percentage to unacceptable levels .

Comment on this story (membership required)
Log in to receive notifications when someone posts a comment

Related blogs

Create a blog about this story (membership required)

Related stories

26 September, 2012
03 September, 2012
23 August, 2012
24 April, 2012
26 March, 2012
29 February, 2012
24 February, 2012
21 February, 2012
13 October, 2011
12 September, 2011
02 March, 2011
18 February, 2011
24 January, 2011
10 December, 2010
12 October, 2010

Featured job

to £90k base, double OTE
London, UK

Find your next job