Charles Cassar and Laura Spiteri, ‘Malta Algorithmic Trading’ (Online Law Journal, 26 February 2015).
DEFINITION
Algorithmic trading concerns trade in financial instruments where a computer algorithm determines individual parameters of orders with little or no human intervention.1 High-frequency algorithmic trading is a sub-category of algorithmic trading concerned with a technique characterised by infrastructure intended to reduce network latencies, a system-determination of order initiation, generation, routing or execution, and high message intraday rates.
MiFID 2 AND REGISTRATION REQUIREMENTS
On the 15th of April 2014, the European Parliament in plenary session adopted updated rules for markets in financial instruments, MiFID 2, Directive 2014/65/EU. MiFID 2 focuses on algorithmic trading activity and is composed of regulations intended to govern this sector. The Directive is not in force yet, but is expected to impact the European securities market in a profound manner.2 The Directive will have a number of implications for algorithmic trading. Algorithmic traders must be registered as an investment firm and all algorithmic trading firms will also need to be registered in accordance with the provisions of MiFID 2, and consequently subject to regulation as investment firms. The Directive moreover requires that algorithmic trading firms provide annual reports to regulators proving that proper risk controls are in place.3
ORGANISATIONAL REQUIREMENTS
The investment firm is to ensure that it complies with the applicable obligations under MiFID and other relevant EU and national laws. Compliance staff should be responsible for providing clarity about the firm’s obligations.
An investment firm that conducts algorithmic trading in one of the twenty-eight EU Member States must notify this to the competent authority of its Member State, the national competent authority being the Malta Financial Services Authority.4 The national competent authority may require the investment firm to provide, on a regular or ad-hoc basis, a description of the nature of its algorithmic trading strategies.5
RISKS AND ALGORITHMIC TRADING
An investment firm engaged in algorithmic trading is to have efficient systems and risk controls suitable for the business it operates. Risks arising from algorithmic trading should be regulated and effective risk controls should ensure that the trading system cannot be used for a purpose which falls foul of Regulation (EU) No 596/2014.6
To enable competent authorities to take sufficient and timely measures against defective algorithmic strategies, it is necessary to flag all orders generated by algorithmic trading. Flagging would allow the competent authorities to react effectively against abusive algorithmic trading strategies which, consequently, pose risks to the orderly functioning of the market.
ESMA AND RISK CONTROLS
European Securities and Markets Authority (ESMA)’s goal is to set the minimum requirements that all investment firms should meet in relation to their trading systems linked to algorithmic trading. It recognises that risks stemming from algorithmic trading are not homogenous across all firms and ESMA thus deemed it appropriate to preserve the proportionality principle established in its guidelines. This concerns allowing market participants to adapt requirements in accordance with the ‘nature, scale and complexity of their business’.7 ESMA laid down that since investment firms themselves should ensure that their systems are fully tested, they should thus remain responsible for making any necessary changes to the algorithmic trading strategy.8
AFME AND THE EUROPEAN COMMISSION
The position of the Association of Financial Markets in Europe (AFME) concerning the European Commission’s proposals to regulate high-frequency trading is that algorithms should operate in a controlled environment with strict controls. AFME suggests that, in addressing concerns of potential market abuse concerning algorithmic trading, the Market Abuse Directive and Regulation are more appropriate than MiFID.9
The European Commission took into account that, although the trading technology has provided a wide range of benefits to the market, it has also presented a number of risks, namely the risk of algorithmic trading generating duplicative or erroneous orders. It has thus suggested a number of regulatory measures.10 The Commission has proposed that markets should temporarily stop trading if there is significant price movement in a financial instrument within a short period of time.
MONITORING
Investment firms engaging in algorithmic trading should monitor the activities of individuals trading on behalf of the firm and the trading activities of their clients.11 If it is the investment firm which provides direct electronic access, it shall be responsible to ensure that the clients using that service comply with the requirements laid down in MiFID 2. The firm shall monitor the transactions to identify any infringements of those rules, disorderly trading conditions, or conduct that may involve market abuse.12
ESMA mentions the need for a ‘kill-button’. This would allow firms to pull all outstanding orders from the market. Given the risks to which algorithmic trading firms are exposed, an effective ‘kill-button’ procedure would ensure adequate risk management and would safeguard the way in which the market functions.13
DUE DILIGENCE
The investment firm should run a validation process of all algorithms and establish a report for the senior management on the basis of a respective risk assessment of the system and algorithms at least twice yearly.14 ESMA’s view is that all prospective and current participants should be subject to adequate due diligence.
RECORD-KEEPING
Investment firms should also keep records about the algorithms used and a description of employed trading strategies. Records must be detailed enough so as to allow the competent authority to monitor the firm’s compliance with the requirements laid down in the Directive.15 Investment firms are bound to keep records of their electronic trading systems for five years.
If an investment firm engages in a high-frequency algorithmic trading, it shall store accurate and time sequenced records of all its placed orders, including cancellations of orders, executed orders and quotations on trading venues. The firm shall make such records available to the competent authority upon request.16
References: 1 Directive 2014/65/EU, Article 4(39) 2 ‘The Impact of MiFID 2 on Algorithmic Traders’ (Chetcuti Cauchi, 26 April 2014) <http://www.ccmalta.com/publications/the-impact-of-mifid-2-on-algorithmic-traders> accessed 15 October 2014 3 PJ Di Giammarino, ‘ESMA gets an earful from industry’ (Automated Trader, 10 July 2014) <http://www.automatedtrader.net/headlines/150856/esma-gets-an-earful-from-industry> accessed 15 October 2014 4 Directive 2014/65/EU, Article 17(2) 5 European Securities and Markets Authority, Guidelines: Systems and controls in an automated trading environment for trading platforms, investment forms and competent authorities (ESMA/2012/122, 24 February 2012) <http://www.esma.europa.eu/mt/system/files/esma_2012_122_en.pdf> accessed 15 October 2014 6 Directive 2014/65/EU, Article 17 7 European Securities and Markets Authority, Discussion Paper: MiFID II/ MiFIR (ESMA/2014/548, 22 May 2014) <http://www.esma.europa.eu/system/files/2014-548_discussion_paper_mifid-mifir.pdf> accessed 15 October 2014 8 Ibid 9 Association for Financial Markets in Europe, Position Paper MiFID 2 – Algorithmic & High Frequency Trading (2012)<http://www.afme.eu/WorkArea/DownloadAsset.aspx?id=6055.> accessed 15 October 2014 10 Ibid 11 Ibid 12 Directive 2014/65/EU, Article 17 13 European Securities and Markets Authority (n 7) 14 Ibid 15 Directive 2014/65/EU, Article 17 16 European Securities and Markets Authority (n 7)
