On 12 February 2018, the FCA published a report on the supervision of algorithmic trading in wholesale markets. The report focuses on 5 key areas of both good and bad practises observed during cross-firm reviews performed prior to the MiFID II enforcement date:
- Defining Algorithmic Trading;
- Developing and Testing Process;
- Risk Controls;
- Governance and Oversight;
- Market Conduct.
Stay current on your favourite topics
The report states that MiFID II requirements were considered when these 5 focus areas were identified, and includes references to relevant MiFID II regulations throughout. It is not unreasonable then, to anticipate that this report, and in particular the practises it identifies, might be used to assess MiFID II compliance against RTS 6 alongside the organisational requirements for investment firms engaged in algorithmic trading. In issuing this paper, the FCA may be looking for a wider set of best and observed best practise that will form the baseline of future RTS compliance assessments.
This is particularly timely, as risk management functions within algo trading firms look to review, evaluate and validate the following as part of their first annual self-assessment for RTS 6 Article 9:
- Algorithmic trading systems, trading algorithms and algorithmic trading strategies;
- Governance, accountability and approval framework;
- Business continuity arrangement;
- Overall compliance with Article 17 of Directive 2014/65/EU, having regard to the nature, scale and complexity of its business.
The timing of the report is, in itself, interesting. This guidance could be seen as a reminder to firms who have deprioritised MiFID II algo obligations that RTS 6 remains a focus area for the FCA as well as regulators in other jurisdictions.
The report’s focus on the governance, process and control aspects of algorithmic trading is an indication to algo trading firms where the FCA thinks their focus should be. The FCA report does not cover, in detail, any best practise for implementing technology to support regulatory compliance. Firms should use the FCA report and its emphasis on non-technical aspects to help drive technological implementations to achieve compliance.
However, there are aspects of the FCA report that will still leave firms needing guidance. For example, RTS 6 Article 10 covers stress testing of algorithmic trading systems and is mentioned only briefly as contributing to the annual self-assessment and validation process. Guidance on running high messaging volume and trade volume tests would have been welcomed as this is an area that is open for interpretation in the regulations. An example of good practise that could have been included is the use of multiple ‘peak’ measures when application development teams are designing high messaging and trade volume tests. The use of a sustained rate – the total peak recorded over a day, combined with a burst rate – the peak volume seen over a shortened period, e.g. intra-second, depending on the applications purpose, improves the level of accuracy of testing for a multitude of stressed market scenarios.
It is important to note that the report highlights a (non-exhaustive) list of examples of poor practise witnessed while conducting the review and it would be prudent for algo firms to focus on these this year:
- Some firms lacked a suitable process to identify algorithmic trading across their business and did not have appropriate documentation in place to demonstrate suitable development and testing procedures are maintained;
- Some lacked a robust and comprehensive governance framework;
- More work needed to identify and reduce potential conduct risks created by their algorithmic trading strategies. This includes delivering suitable market abuse training for staff involved in the development and implementation processes;
- Firms also need to consider the potential impact their algorithmic trading activity (including the combined impact of multiple algorithmic strategies) may have on the fair and effective operation of financial markets.
Stay current on your favourite topics
Firms must keep focused on compliance and consider what regulators will look for beyond the text in RTS 6 (summary here). It is right to assume that the FCA report forms only part of the regulator’s baseline expectation when assessing compliance with RTS 6. It should be one of multiple inputs used by an algo trading firms risk management function for their annual self-assessment and validation process. Any algorithmic trading self-assessment must be defensible in front of regulators. This should begin with a thorough examination of the criteria set out in in RTS 6 Annex I.