The FX industry has been caught out by the advance of technology before, so although the report looks at the issue across broader financial markets, the FCA’s study on algorithmic trading should be essential reading for anyone senior in the industry. What certainly concerned me reading the report was further evidence of a lack of real understanding about the potential impact and risks associated with using algos at the highest level. Of course, whilst trying to discuss and highlight some of these themes I have not missed out on the opportunity to suggest names for potential algo strategies!
The report lists some good and poor practices the FCA identified as part of the study and almost universally the poor practices exhibit signs of a “near enough is good enough” attitude, as well more general ignorance. Neither can be acceptable in this day and age when more and more algos are being deployed.
Although, inevitably, MiFID II plays a role in defining the oversight framework, the FCA report focuses mainly on five key areas; definition, development and testing, risk controls, governance and market conduct.
When it comes to definition of algo trading, it found that some firms apply a too generic approach, which means less high profile areas of the business that may be using algos, are often overlooked. Partly this is the result of the increased noise around one of 2018’s early buzzwords – algo – which has led some firms, who are scared of being left behind, to roll out algo technology without really thinking about it.
It is a sad fact of life, but a sales person or division head hates to have to say to the wider world they don’t have something their peers do and this can lead to the rushed release of algo products. From personal experience it is not so much the lack of technical controls that worries me, more that occasionally the person talking to me about algos understands little or nothing about them – and yes, it is that obvious! – and that is a real lack of control.
Incredibly there are firms out there who do not formally log incidents or breaches involving their algos, according to the FCA. This strikes me as being not only incredibly lazy and stupid, but dangerous for both the firm concerned and the wider market. Again, this highlights the lack of preparedness and a rush to market,
These people obviously also get off on having exotic or exciting sounding names for their algos, but perhaps they should all call them by one – FOMO?
The FCA report also finds that some firms conduct reviews of their algos’ effectiveness and appropriateness on an ad hoc basis and have no formal process for identifying material or substantial changes. This may just be a question of timing and the work is underway – certainly the FCA will have informed these firms of their apparent shortcomings – but allied to another finding, a weakness around risk controls that also work too generically (the FCA believes they should work in smaller time pockets – such as 15 minutes), the picture is one of an institution hoping something doesn’t go wrong. This is one for the Brits but these firms should call their strategy, Nelson.
That name could also cover those firms that the FCA found only have high-level, generic descriptions for their algos, strategies and systems.
The report also finds that some firms have different development and testing standards across business lines, something it feels is poor practice, although there is a case to argue that the different market structures encountered may naturally lead to this evolution. There should certainly be minimum testing standards, this is something I have discussed before and would refer readers to the report published by EcoFinancial Technology a couple of years ago.
Some firms also lack checks and balances surrounding how algos are functioning and do not provide sufficient information to senior management to allow them to provide adequate training and testing environment.
Collectively, this “near enough is good enough” approach should result in a new algo strategy entitled WIBI – Win it or Bin it – an expression widely, and topically, used by athletes taking part in the luge, skeleton bob and other ridiculously stupid sports that involve going down the Cresta Run on a tea tray.
Moving on – and yes I am going to grind through this to the end – the FCA also found that some firms failed to keep adequate documentation around the development, testing and deployment of algos and states, “As such, a number of key decisions are taken without an adequate audit trail and without sufficient explanation of the underlying rationale for these decisions.”
Obviously, this algo strategy needs to be entitled “Man” although there could be some bespoke work done which turns it into a “Donald”.
Lastly, and this is more serious and something I have come across repeatedly, including a recent telling of a story of a senior FX manager who didn’t understand latency, the FCA report finds firms, “Where senior managers are not able to demonstrate the required knowledge to be able to provide effective challenge to front line algorithmic trading operations.”
The report also notes how compliance functions lack the required knowledge to be able to challenge the front office when it comes to algo deployment and performance.
The banking industry in particular has suffered (and continues to suffer) years of being fined for a lack of oversight and understanding over one of the more simpler developments of the last decade, social media and in particular chat rooms. The broader industry has also witnessed events involving algorithms, most famously I suppose at Waddell Read and Knight Capital, where it was pretty obvious that no-one knew what to do (or even what was happening probably).
It is all very well me making up names for algo strategies (feel free to send in your own of course – my favourite is a friend of mine who proposed the institution call their new algo Ralph – no acronym, just Ralph) but there is a serious point here. The past few years have seen what I consider a disturbing trend where the buck has stopped some way short of the top of the business. A lot of good people have been thrown under the bus by senior management and somehow, it has been mysteriously decided to accept this.
It used to be the buck went all the way to the top – now apparently it stops on the fourth floor. Reading this FCA report it concerns me that if something does happen, then again, it will be someone relatively insignificant that cops the blame because in some firms management are jumping on a bandwagon, using algos as a vanity project or turning a blind eye to real risks.
You’d like to think that the industry has learned a lesson from the past decade, but perhaps it hasn’t.