By Peter Lawrey, CEO at Chronicle Software.
Regulations like MiFID II and the Market Abuse Regulation (MAR) in Europe, and Dodd Frank in the US, have raised the bar with respect to trade surveillance as regulators seek to boost transparency and investor protections across the board. This has created the need for financial institutions to capture and store vast amounts of data without impacting the performance of their trading mechanisms. Chronicle Software has worked with investment banks and hedge funds to deliver trading systems capable of capturing and storing the large data sets required to achieve and maintain compliance.
In response to the new regulatory stress on surveillance, trading firms’ front-office operations are being forced to adopt more proactive processes and systems to monitor the activities of their staff and communications with the marketplace. But even as regulators seek to impose a firmer grip on trading markets in the hopes of weeding out unscrupulous behaviours, liquidity venues and electronic communications mechanisms are fragmenting. Trading conversations – with execution engines, clients and counterparties – are conducted through an array of connections.
These have evolved from simple fixed-line voice systems to mobile devices and a growing number of official and unofficial electronic communications channels. Today, conversations with a client in the run-up to a trade may start from the trading desk phone line, then move to Bloomberg chat, WhatsApp, and end with the final decision delivered via mobile phone.
Regulations require financial firms to monitor these real and electronic conversations in order to capture any and all information that may be material to the final transaction decision. MiFID II, for example, obliges firms to capture and retain all communications that could relate to a transaction, whether or not the transaction actually takes place.
According to Article 16(7) of the regulation, firms must record and story “telephone conversations or electronic communications relating to transactions concluded when dealing on own account and the provision of client order services that relate to the reception, transmission and execution of orders … even if those conversations or communications do not result in the conclusion of such transactions”.
This requirement applies not only to the traditional fixed lines but also to mobile phones, SMS (text messaging) and any phone- or tablet-based instant or other allowed messaging application used in trading-related conversations, such as chat, email, Bloomberg Mail/Messaging, Skype and WhatsApp.
To meet the surveillance obligations embodied in MiFID II, MAD/MAR and Dodd Frank, firms need to ensure comprehensive data capture together with traceability and reproducibility without negatively impacting performance.
Key to compliance is the storage of trading records, and archiving appropriately of these records to ensure ease of access in response to a regulatory enquiry. European regulator ESMA says “firms are required to keep records produced under Article 16(7) of MiFID II for five years, with the extension to seven years, if requested by the competent authority.”
Retrieval must be fast, precise and complete; all records must be immutable/durable, meaning that records must remain intact once they are recorded. Finally, MiFID II specifies a time-stamp granularity for voice-based trading of one second, with maximum divergence from the benchmark Coordinated Universal Time (UTC) of one second.
These requirements present challenges with respect to accessing the complete set of records for trade reconstruction as evidence for regulatory inquiries. Getting data capture right is essential. Poor quality or insufficient data to compare transactions can result in trade surveillance activities that generate a significant amount of false positives, creating more work for surveillance teams as they sift through bad signals looking for an accurate indicator of improper behaviour.
Existing trade surveillance systems also often generate false positives through their inability to draw upon the whole data set required to construct a clear picture of market situations at the time of trade. What’s needed is the ability to investigate market abuse using the whole reference data set for each instrument, which can reduce false positives by pinpointing the precise security in question and eliminating other potential suspects identified by smaller reference data sets. Flagging transactions on the basis of company name alone, for instance, can create false positives by confusing different legal entities within a corporate hierarchy, or entirely different companies with similar names.
An effective market surveillance solution needs access to the whole of the firm’s trade data. Many commercial solutions struggle to ingest data from the wide variety of trading platforms a firm may be using, undermining efforts to maintain data quality and introducing false positives. It also needs the ability to apply additional data – such as instrument volume or volatility – to help reduce false positives.
To avoid the pitfalls of many existing surveillance solutions, firms need to underpin their monitoring efforts with a deterministic system that captures all data input into a trading application but also reads all output data. The platform also should combine full audit capability with latency monitoring and robust failover.
This approach can give firms the ability to record everything the trading application is doing. Few available systems are able to deliver this traceability without impacting the functionality of the application. Most are able to record output only. By implementing a deterministic system that eliminates duplicates and reduces redundancy of information, firms are able to ensure complete auditability with no performance loss.
The benefits of taking this approach go beyond traceability, reproducibility and auditability. Firms are able to reduce time to market for implementing surveillance processes or indeed any regulatory imperative involving data capture and audit.
By design, a platform of this type won’t function if data is missing. In this way, firms can have complete confidence in the data being processed, whatever the volume of throughput, eliminating the need for analysts to go back and fill gaps where data is missing, which is expensive and difficult, if not impossible.
This deterministic approach gives firms a full picture of all data instances relating to all transactions. To minimise false positives, firms need detailed logging of data relating to all possible scenarios, potentially requiring examination of vast quantities of data.
At Chronicle, we’ve tested throughputs of our deterministic messaging platform of bursts of 128 GB of data at over 100 million messages per second. This gives our clients confidence not only in the traceability, reproducibility and auditability of their data to meet the surveillance requirements of MiFID II, MAR and Dodd Frank.
If you are dealing with millions of updates per second, and need to get your systems in order to meet your market surveillance or other regulatory audit obligations, you should be talking to us.
