Bryok Blog

  • Home
    Home This is where you can find all the blog posts throughout the site.
  • Categories
    Categories Displays a list of categories from this blog.
  • Tags
    Tags Displays a list of tags that has been used in the blog.
  • Bloggers
    Bloggers Search for your favorite blogger from this site.
  • Login
Steve Leegood

Steve Leegood

Steve Leegood has not set their biography yet

Posted by on in Market Integrity

Foresight: The Future of Computer Trading in Financial Markets
- A Critical Review

How many leading academics and experts does it take to change a light bulb?

The Summary[1] of the final report of Foresight: The Future of Computer Trading in Financial Markets (2012)[2] suggests that 150 is not the right number. What is a light bulb? Why does it need to be changed? What does it need to be changed into? Maybe it's the light switch that needs to be changed? Is the lack of light really a problem? Research shows that blundering around in another room looking for something else does not reveal a problem with the bulb. Why can't these pesky humans put up with being kept in the dark anyway?

Our comments here refer to the Summary. We will address the full Report separately.

Given the time, money and sheer number of international academics involved in this Foresight Project, its conclusions and recommendations are banal and embarrassingly thin.

The key message at the head of the Executive Summary states "… further work is needed to inform policies in the longer term, particularly in view of likely uncertainties and lack of data. This will be vital to support evidence-based regulation ...". Well, indeed, and was not the point of the Project to produce this evidence? But what is a "likely uncertainty"? Is that like peer-reviewed woolly thinking?

As far "lack of data", this is pathetic. There is no lack of data; it is created by the terabyte by the markets every day (we actually provided some to the Project). It is simply untrue to say, as the Summary does, that "data of the quality and detail required to identify abuse are simply not available to researchers". It is available, and our own research using it reveals extensive market abuse.

Define Your Terms

The first aim of the Project was to look into "computer-based trading (CBT) in financial markets". Its focus is on "HFT and AT" (High-Frequency Trading and Algorithmic Trading). Like Fish'n'Chips. These two distinct concepts are not defined properly or differentiated. A better distinction might be between investing and trading.

Investing is what society needs, and what securities markets originally existed to support. It is inherently a relatively long-term activity.

Trading, though, is a zero sum game. It may enhance the efficiency of markets, to the benefit of investors and society, or it may be predatory - seeking to win the game at the expense of others, by fair means or foul. Or both. Most of this activity is "computer-based", much is automated.

And algorithmic trading might be rapid, but that does not make it high-frequency. HFT is really about opening and closing positions very rapidly with the aim of making money from short-term or cross market price differences (and some might say to create the differences in the first place in order to profit from them).

These terms do not provide a useful definition for the problem to be investigated. A better question would be "is HFT, defined in this way, a benefit to society, or does it damage society; in what circumstances; how could the benefits be enhanced and the damage limited or prevented?"

The second, and very different, question would be "if a trading decision is automated, what happens in extreme or unexpected situations; what happens if two trading automata interact with other unstably; how could the impact on market integrity best be minimised or removed?"

This failure to define terms and ask the right questions fatally wounds the Project before it has even begun. Little wonder it leaves the real questions unanswered, and pleads instead for "further research", involving the "wider academic community" - the luminaries of which have made such a hash of this Project. This confusion pervades the Summary.

Answer The Question

Having couched its problem as a technological one, the Project is forced to draw conclusions that might be best characterised as "the bleeding obvious" - "new technologies will continue to have profound effects on markets"; "there will be increasing availability of substantially cheaper computing power" . Well, duh!

The technology itself is irrelevant, it is the uses to which it is put, and the motives for that use. A faster box only helps a cheat to cheat faster: the problem is not the box, it is the cheat, and the system that provides the opportunities, or even justification, for his cheating.

King Canute Redux

The danger of blaming the technology is apparent in the conclusion "increasing use of computers and information technology in financial systems is likely to make them more, rather than less complex …" (true, if trite), so "… constraining and reducing such complexity will be a key challenge for policy makers." But it is simply impossible to constrain this complexity - it is inherent in the distributed nature of the markets, and of course of the technology they run on.

Rather we need to recognise this impossibility and instead deal with the unfair exploitation of the advantages available to some participants, and the disorderliness that may be created by the overall markets system itself. It is the integrity of the markets - real and perceived - not their complexity which is the real issue. We therefore do agree with conclusions on "adequate dissemination and storage of trading data" and "making sense of disclosed information and developing a better understanding of the financial system", and on "accurate, high resolution, synchronised timestamps".

However, that this "implies the need to harness the efforts of researchers" is not obvious, even if it may be self-serving, if they're the same researchers who have contributed to the Project's failure effectively to understand the financial system.

Understand Cause And Effect

CBT has reportedly benefited markets, eg by improving liquidity "as measured by bid-ask spreads". Is that how you measure liquidity? If you do not define liquidity you are not equipped to examine deleterious effects such as "periodic illiquidity", for example. (And is it really "periodic", or do they mean occasional, or unreliable, for instance.)

Falls in transaction costs are reportedly "related closely to the development of HFT". Correlation is not causation, anyone? Did we forget that MiFID introduced competition for instance. "Market prices have become more efficient", what does this even mean?

The "three main mechanisms that may lead to instabilities" are another statement of the obvious. Which of the vaunted 50 peer-reviewed academic studies came up with this pabulum?

I'm not sure whether I more dislike the incompetence, ignorance or the illiteracy.

Out Of The Mouths Of Babes…

Just when an astute and tenacious reader would be about to conclude that this is an unremittingly negative diatribe, we do have something positive to commend. The Summary, albeit somewhat circumlocutorily, does correctly observe that HFT causes a significant perceptual issue for the markets. Society needs investment markets to be fair and orderly, and to be perceived to be so. Otherwise confidence and participation in the markets are damaged.

However, it goes on to say "regulators and policy makers can influence perceptions, even if definitive evidence on the extent of abuse will not be available to settle the debate". How does that make sense? The regulators should say "look, we don't know whether there's a problem or not, despite the extensive Foresight Project, but don't worry your pretty little heads, it's only your perception that's at fault"?

Why not put the onus on the cause of the perceptual problem: the HFTs. Prove you're beneficial, prove you're not abusive, prove you cannot cause instability or "endogenous" risks. And I don't mean turning up at conferences or workshops with spurious arguments or irrelevant "evidence", I mean prove it. Otherwise you can whistle for your trading licence.

Start Making Sense

While we cannot agree in general with the Summary conclusions - since we think are they based on wrongly defined and poorly researched premises - some are sensible or unobjectionable.

In particular we support the idea "to learn lessons from other safety-critical industries, and to use these to inform the effective management of systemic risk in financial systems". Indeed we have undertaken our own research into this. We do not agree that this should be "in the longer term", but would be useful right now.

We also support "making surveillance of financial markets easier". However the specific conclusion; "development of software for automated forensic analysis of adverse/extreme market events", seems to have come from leftfield. As useful as this would doubtless be, surely regulators first need data and tools to analyse it properly. They need fully detailed, timestamped orderbook event data (not just trades, and not just on request), allowing each actor in the order lifecycle to be identified. They need systems able to accept, process, store and retrieve this data - in the order of a billion events daily. They need tools to visualise the data, and to perform advanced mathematical and statistical analysis of it in multiple dimensions. We have also researched such tools.


[1] www.bis.gov.uk/assets/foresight/docs/computer-trading/12-1087-future-of-computer-trading-in-financial-markets-summary.pdf

[2] www.bis.gov.uk/assets/foresight/docs/computer-trading/12-1086-future-of-computer-trading-in-financial-markets-report.pdf

Hits: 4975
0

© Bryok Systems Limited 2013

200 OK

OK

The server encountered an internal error or misconfiguration and was unable to complete your request.

Please contact the server administrator, [no address given] and inform them of the time the error occurred, and anything you might have done that may have caused the error.

More information about this error may be available in the server error log.