Thursday, 25 October 2012

The Future of Computer Based Trading

The UK Government has published its Foresight Report on the Future of Computer  Trading in the Financial Markets .  "Foresight" is a bit of a mis-nomer here, given that Computer Based Trading (CBT) has already had a profound effect on the markets across the globe.

What first struck me about this Foresight report is that is presents particular solutions that have undergone a cost-benefit analysis. BIS Foresight reports typically highlight problems and the recommendations can usually be summed up as "the issue is complex, policy makers need to initiate inter-disciplinary research".  Also, the Project Team was large and the Acknowledgements list long.

I interpret this as reflecting a fact, that people do not really know what is going on.  Recent reports from BIS (covering Migration, Climate Change, Food and Farming) all relate to large, well established academic fields.  At the time the Foresight report was initiated I do not think there was a well established research centre in CBT in the UK, though UCL and Essex are now building them.

The report does recognise this, with John Beddington's summary noting that

there is a relative paucity of evidence and analysis to inform new  regulations, not least because of the time lag between rapid technological  developments and research into their effects. This latter point is of particular concern, since good regulation clearly needs to be founded on good evidence and sound analysis.
 This is developed on p 11
Markets are already ‘socio-technical’ systems, combining human and robot participants. Understanding and managing these systems to prevent undesirable behaviour in both humans and robots will be key to ensuring effective regulation: While this Report demonstrates that there has been some progress in developing a better understanding of markets as socio-technical systems, greater effort is needed in the longer term. This would involve an integrated approach combining social sciences, economics, finance and computer science. As such, it has significant implications for future research priorities.
My principal concern is that this important point, directly relevant to BIS, will be lost amid the specific recommendations in the Report.  BIS might be excused in offering solutions given that the problems are current, however I doubt that BIS has the ability to address the issues, either in terms of skills or authority.

The Report is useful in that it does categorise different types of CBT.  I should declare an interest: my mathematical research is in optimal control, and this is relevant to the optimal execution of large trades, which  is described as 'agency' based, as opposed to proprietary, trading.  This is a very active area of mathematical research, though I am not convinced there are robust practical solutions.  The key issue is that current (optimal stochastic control) algorithms make assumptions on the trading environment, and implicit is that the environment is static (ergodic if you prefer).

However, I feel the Report could be clearer about distinguishing agency and proprietary trading.  This distinction was traditionally a part of the London markets, that disappeared with Big Bang in the 1980s.  In the seventeenth century John Houghton describes stock-brokering as
 the Monied Man goes amongst the Brokers (which are chiefly upon the Exchange [Alley], and at Jonathan's Coffee House [the origins of the London Stock Exchange], sometimes at Garraway's and at some other Coffee Houses) and asks how Stocks go? and upon Information, bids the Broker to buy  or sell so many Shares of such and such Stocks if he can, at such and such Prizes
Brokering has always been seen as reputable, unlike stock-jobbing or dealing, which was described by  Daniel Defoe in 1719 (the year he published Robinson Crusoe) in an article  The Anatomy of Exchange Alley  as
a trade founded in fraud, born of deceit, and nourished by trick, cheat, wheedle, forgeries, falsehoods, and all sorts of delusions; coining false news, this way good, this way bad; whispering imaginary terrors, frights hopes, expectations, and then preying upon the weakness of those whose imaginations they have wrought upon

Thomas Mortimer described  in 1761 the type of person involved in stock-jobbing; there are three types of stock-jobber, firstly foreigners, secondly gentry, merchants and tradesmen, and finally, and "by far the greatest number'', people
 with very little, and often, no property at all in the funds, who job in them on credit, and transact more business in several government securities in one hour, without having a shilling of property in any of them, than the real proprietors of thousand transact in several years
Modern criticism of high frequency trading  is a continuation of a long tradition of distinguishing the activities of Monied Men investing in the market and Traders speculating within the market.  As such it can get entangled in a web of social judgements, such as it is OK for the rich to gamble but not the poor.  To make decisions about HFT, therefore, requires some understanding of this mess.

I regularly make the point that finance is not just an important social utility, but a melting point for scientific innovation.  One lesson that science, as a whole, can learn from contemporary finance is the problem of "calculability".  Nineteenth century science relied on this principle, and twentieth century science, to a large degree, accepted problems of chaos and complexity, but still relies on determinism, which is often missing  in the social sciences. The assumption of "calculability", that induction is possible, is at the root of many of the debates about science and, for example, the climate debate is not about the raw data but the models that process that data.  This is a broad issue and relates the the UK's science minister's recent comments on how "red tape" gets in the way of progress. The Minister argues that in Europe there is
an approach to regulation so far removed from any rational appraisal of risk that it threatens to exclude Europe from many of the key technologies of the 21st century
The Science Minister cites examples from the physical sciences, I wonder if he is so confident in the value of invention in finance?  I often worry about terms like "rational appraisal of risk" because I feel all to often it is based on assumptions of stable chances.

The Report echoes the views of Willets when it opens with the ambiguous statement

A key message: despite commonly held negative perceptions, the available evidence indicates that high frequency trading (HFT) and algorithmic trading (AT) may have several beneficial effects on markets. However, HFT/AT may cause instabilities in financial markets in specific circumstances. (my bold)

The sense appears to be "we don't really know, but hey-ho lets see what happens".   This makes me very nervous.

The main issue is, predictably, related to HFT.  HFT is "good" in that it enables price-discovery.  It is "bad" in that it might destroy liquidity in times of stress.  So, generally good, occasionally really bad.  However this is based on an assumption tied up with the whole "efficient markets" paradigm,

pricing is efficient when an asset’s price reflects the true underlying value of an asset; (Note 4 p 19)
This is a Platonic argument implying a problem of epistemology:  mere humans have difficulty in perceiving the Real price, but by having HFT we get closer to Reality.  If you reject this, and believe that there is an ontological perspective: a "true" price may not exist, the justification of "generally good" falls away.  My work on Ethics and Finance: The Role of Mathematics is concerned with this issue.

Finance is critical to society, as we have all found out recently.  We should be debating finance with the vigour and passion that we debate climate change, GMOs, nano-technology, energy and so on.  Without this debate we cannot establish the values upon which the evidence base, which informs the regulations, is built.  I feel that the Report misses making this point clearly, and so cannot really address the issue in hand.

Andy Haldane, the man at the Bank of England responsible for financial stability was initially involved in the Foresight project but withdrew.  Haldane has recently discussed the issue complexity in financial regulation.  He appears to reject the premise that finance is susceptible to calculation and rules and advocates judgement, a virtue ethics argument.  I wonder if we should trust the Bank of England more than the Department of Business Innovation and Skills on the issue of computer trading.

No comments:

Post a Comment