2018年-FCA英国金融行为监管局_algorithmic_trading_compliance_wholesale_markets_28页_350kb

Algorithmic Trading Compliance in Wholesale Markets Summary

Executive Summary

This report outlines the key areas of focus for algorithmic trading compliance in wholesale markets, particularly under MiFID II. It emphasizes the importance of effective systems and controls, governance, risk management, and market conduct. The FCA and other global regulators are increasingly concerned with the risks associated with algorithmic trading, especially as it becomes more prevalent and complex. The report provides examples of good and poor practices observed in firms, highlighting the need for consistent and well-documented processes across all stages of algorithmic trading.

Key Areas of Focus

• Defining algorithmic trading: Firms must establish a clear process to identify algorithmic trading, manage material changes, and maintain a comprehensive inventory.
• Development and testing: Robust, consistent, and well-understood development and testing processes are essential to identify potential issues before deployment.
• Risk controls: Pre- and post-trade controls must be suitable and robust to monitor, identify, and reduce trading risks.
• Governance and oversight: An appropriate governance and oversight framework is required to ensure effective challenge from senior management, risk management, and compliance.
• Market conduct: Firms must consider the impact of their algorithmic trading on market integrity, monitor for conduct issues, and reduce market abuse risks.

Defining Algorithmic Trading

Core Definition

Algorithmic trading is defined as trading in financial instruments where a computer algorithm automatically determines individual parameters of orders (such as whether to initiate, the timing, price, or quantity of the order, or how to manage the order after submission) with limited or no human intervention. This excludes systems solely for order routing or post-trade processing.

Good Practice

• Conduct extensive reviews across the business to define and capture all algorithmic trading strategies.
• Maintain detailed policies and training for staff to consistently apply definitions and controls.

Poor Practice

• Apply a high-level definition without considering the business's specific activities.
• Conduct basic identification exercises that miss less frequent algorithmic uses.

Substantial or Material Changes

Key Requirement

Firms must have a process to define and identify substantial or material changes to their algorithms, strategies, or systems. These changes require further testing and record-keeping.

Good Practice

• Maintain well-defined policies with detailed criteria for identifying changes.
• Include suitable training for all relevant staff to ensure consistent application.

Poor Practice

• Conduct ad-hoc reviews without a formal process.
• Fail to demonstrate a consistent and well-understood methodology for identifying changes.

Algorithm Inventory

Key Requirement

Firms need to establish and maintain a comprehensive inventory of algorithmic trading strategies and systems, including objectives, development procedures, testing, owners, users, and risk mitigants.

Good Practice

• Retain detailed documentation on the types of algorithms, strategies, and systems, including operational objectives, parameters, and risk controls.
• Ensure all key decisions are recorded with an audit trail.

Poor Practice

• Lack clearly defined inventories.
• Provide limited documentation, resulting in poor audit trails and unclear rationale for decisions.

Development and Testing

Key Objective

To ensure firms maintain robust, consistent, and well-understood development and testing processes to identify potential issues before deployment.

Good Practice

• Use a phased development process with independent checks and balances.
• Encourage open communication and have a separate team for verification.
• Document all stages and include independent sign-off from relevant functions.

Poor Practice

• Apply inconsistent methodologies across different trading desks or business lines.
• Use a simplistic final sign-off process with no independent representation.

Deployment Procedure

Key Requirement

Firms must establish a suitable and controlled procedure for deploying new or updated algorithms into live environments.

Good Practice

• Have detailed staging and scheduling plans.
• Implement procedures to identify and roll back issues during deployment.

Poor Practice

• Lack coordination between front-line and support/control functions.
• Fail to provide effective management information on conformance testing, operational arrangements, and surveillance procedures.

Documentation and Audit Trail

Key Requirement

Firms must ensure adequate documentation and a comprehensive audit trail throughout the development and testing process.

Good Practice

• Maintain comprehensive developmental evidence covering theoretical construction, behavioral characteristics, input data, and code protocols.

Poor Practice

• Retain limited documentation, leading to poor audit trails and unclear rationale for key decisions.

Risk Controls

Key Objective

To ensure firms develop suitable and robust pre- and post-trade controls to monitor, identify, and reduce potential trading risks.

Pre-Trade Controls

• Basic controls: Overall limits applicable throughout the trading day.
• Enhanced controls: Limits split across time periods or per symbol, with dynamic settings based on average daily volume (ADV) and touch size.

Control Setting & Amendment Process

• Controls must be tailored to the type of trading activity and set at appropriate levels.
• Regular review of control settings is essential to ensure they remain suitable.

Good Practice

• Maintain detailed controls at multiple levels (client, strategy, firm-wide).
• Provide oversight by an independent risk function and allow pre-authorised staff to adjust controls with pre-agreed levels.

Poor Practice

• Apply broad controls across multiple clients or strategies without considering activity levels.
• Set control levels significantly higher than normal trading activity, making it hard to demonstrate effectiveness.

Post-Trade Controls & Monitoring

Key Requirement

Effective post-trade monitoring is critical to ensure compliance with MiFID II requirements, including continuous assessment of market and credit risk exposures, accurate trade and account information, and reconciliation with third parties.

Good Practice

• Maintain complete, accurate, and consistent trade and account information.
• Ensure traders and the risk function undertake post-trade monitoring.

Poor Practice

• Fail to maintain accurate and consistent post-trade records.
• Lack effective monitoring procedures, leading to potential compliance failures.

Conclusion

Algorithmic trading compliance in wholesale markets is a complex and evolving area. Firms must ensure they have appropriate systems and controls, maintain clear documentation, and adopt robust governance and oversight frameworks. The FCA continues to supervise and assess firms to ensure they meet these requirements and reduce the risks associated with algorithmic trading.