Best Practices in Algorithmic Trading Compliance

A significant amount of order flow is handled by algorithms nowadays. That’s because algorithms allow firms to make more efficient buy and sell decisions. In addition, algorithms can execute orders far more efficiently and with less market impact – which is especially important in the fragmented market structures in Europe and the US. But algorithms have a downside, too. We’ve experienced enough full-blown and mini flash crashes over the last decade or so to understand just how disruptive a rogue algorithm can be.

The regulators have thought long and hard about how to strike a balance between enabling market efficiency and protecting market integrity. In Europe, MiFID II and the Market Abuse Regulation (MAR) address the obligations and responsibilities firms need to assume if they want to use algorithms. The regulations include additional requirements for firms who engage in algorithmic trading to pursue a market-making strategy. This spans all investment decision or execution algorithms, whether in high frequency trading or elsewhere. Suffice it to say that if you use algorithms, you’re responsible for them – even if you outsource technology.

Every firm is different, so the regulators don’t want to be too prescriptive, but they still expect firms to adhere to certain principles. And if they don’t comply, they’ll expose themselves to serious fines, penalties and possibly even imprisonment. Covering all the bases is no easy feat. So where do you start?

The UK’s Financial Conduct Authority (FCA) recently published Algorithmic Trading Compliance in Wholesale Markets, which lays out a broad framework for structuring a compliance program. The framework comprises several sections including: defining algorithmic trading, development and testing, risk controls, governance and oversight, and market conduct. Each section states an objective, and describes what good practice is and what poor practice is.

Here are some of the best practices outlined in the report:

Review how algorithms are used within the firm. Document the different types of algorithms, trading strategies and systems, including relevant operational objectives, parameters and behavioral characteristics. Break down the various components or algorithms contained within the strategy or system. Identify who “owns” the algorithms and who is approved to operate the strategy or system. Establish policies on the completion of development, validation and testing procedures, along with appropriate sign off from senior management and other relevant control functions.

Establish a robust development and testing process, and appoint a project lead to oversee it. Break down the process into separate phases and establish independent checks and balances at each stage. Conduct due diligence checks to assess potential risk and establish risk control thresholds. Test the systems to see how they perform in extremely volatile conditions. Open communication between different business units is good, but a separate team should verify and checks the output and quality of the code. Establish a formal approval and sign-off process.

Involve senior management in the development and testing process and have compliance do a gap analysis. Executives must understand the potential market implications, and be apprised of the firm’s ability to supervise algorithmic trading activity. It might be necessary to establish new roles/responsibilities to focus on this activity.

Maintain detailed risk controls at multiple levels, review them regularly, and have an independent risk function oversee the process. Firms should have controls on a client and/or trading strategy level depending on the type of business undertaken, as well as on a firm wide basis. Certain pre-authorized front office staff may be allowed to adjust specific controls with levels pre-agreed with the risk function.

Set up dedicated teams to monitor activity.Tailor monitoring and surveillance systems to specific risks within your algorithmic trading activity. Surveillance teams may use visual and audible alerts, including automated control thresholds where alerts are generated at pre-defined levels such as 50% and 80% of a control limit. A committee comprising representatives from the trading, client coverage, compliance, risk and credit teams should conduct regular reviews of control levels.

All relevant members of staff should go through market abuse training. A strategy may not strictly meet the definition of market abuse, but staff needs to know when a strategy could have a negative impact on market integrity or contribute to wider market disruption.

Firms that follow the FCA’s framework will be on the right path to compliance. A key takeaway from the report is that knowing your algorithms involves a mix of teamwork and technology. Fortunately, advanced systems are available to supplement human skills, allow firms to cope with the vast datasets and fast pace of today’s financial markets, and manage risk effectively.

Nasdaq’s risk and surveillance solutions help firms comply with the regulations governing algorithmic trading. Its alerts focus on proactive discovery through intelligent analysis of market data as well as electronic and voice communications. Revealing the intent behind trading activity is of utmost importance. Surveillance professionals can incorporate behavioral analysis by reviewing time series, applying contextualization, identifying emerging patterns and benchmarking them against the norms, and deploying machine learning. Ultimately, these systems are the cornerstone of market integrity and the key to supporting investor confidence.

To learn more contact us at

The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.

More Related Articles

Sign up for Smart Investing to get the latest news, strategies and tips to help you invest smarter.