Meeting Your Best Ex Obligations: How Do You See the Wood for the Trees?

Originally appeared in The Trade.

The past year has been a period of significant change for equity trading as a result of MiFID II and all the structural changes it has introduced. Whether it be from the increased scrutiny over operating procedures through enhanced Best Execution requirements, the introduction of new venue types or changes to the trader’s workflow through increased automation and the use of new tools, all of these have changed the accepted norms and challenged the conventional thinking on what works and what doesn’t.

Therefore, how do we cut through the often conflicting sources of information and data to really see the wood for the trees and understand what’s working and what isn’t?

Why is it important to review?

Before we dive headlong into the analysis, we should understand why it is important to do this analysis, because as is often the case with formidable challenges, it might seem easier to assume everything is fine and to therefore move on to less challenging tasks.

MiFID II extended the requirements of firms to demonstrate Best Execution by ensuring firms monitor their execution counterparties and fully understand their operating procedures. Further to this, the FCA were clear in stating that trading desks should know how their traders are interacting with the market, particularly through algorithmic trading.

To fulfil these requirements, firms should have a very clear framework for the continual evaluation of counterparties that leads to decisions being made on who those counterparties are. This should be completely independent of any other potentially conflicting decisions, such as research provider selection.

A framework for evaluation

So how do we begin to evaluate how a counterparty is handling one’s client’s orders? Through experience of analysis carried out over many years, three key areas of a counterparty’s systems have historically provided the answers: Liquidity, Mechanics and Protection.

Each of these areas has a direct effect on the execution performance achieved: What liquidity is your order interacting with? How is your order being handled and how are liquidity providers being interacted with? What protections are the counterparty using when executing your order?

Analysing liquidity

Liquidity has become more fragmented post-MiFID II with the introduction of new venue types, particularly periodic auctions and systematic internalisers (SIs) that are now accounting for approximately 1.5% and 15% of the EMEA equity market respectively. These new venues offer significant liquidity, however this is spread across multiple providers – six in the case of periodic auctions and over 50 individual SIs. Each provider has a unique liquidity profile, instrument universe, average execution size and toxicity statistics. For that reason, the question is which liquidity sources should I be interacting with? Which are providing ‘good’ or ‘bad’ liquidity?

It is incumbent on the execution counterparty to have a very clear rationale for all the execution venues they interact with on your behalf, and as a result, they should be able to evidence this rationale when asked. If an execution provider is not able to do this, it would be reasonable to question whether they have the appropriate controls and governance framework around their venue on-boarding process for you to be comfortable with trusting that counterparty.

At Liquidnet we analyse a range of different metrics, both at a parent and child order level to evaluate the execution quality a liquidity source is providing. We call this our Venue Ranking Model and it can be tailored for different trading objectives by adjusting the weighting given to each metric. The output of the model is reviewed as part of our Best Execution Governance framework and Smart Order Routing (SOR) decisions are taken as part of that process.

Analysing mechanics

Understanding where your order is being taken by a counterparty gives you a degree of clarity, however the next aspect to understand is how these liquidity sources are being interacted with on your behalf. This is most often associated with Smart Order Router mechanics.

Firm, conditional, RFQ, streaming quotes; all different methods for executing on different venues and many of the venues offer multiple interaction options. Which of these methods is a counterparty using when transmitting your order and interacting with the different liquidity sources? What is the rationale for using one method vs. another? For example, it is common thinking that the optimal way to interact with SI liquidity is by reacting to streaming quotes being consumed by the counterparty. However, in many cases these quote streams are not available and the only option is to send a firm order to the SI operator’s SOR and to allow them to react to a quote on your behalf.

If this is the workflow a counterparty is using, then what information is being passed on to the SI? Is there any form of flow profiling occurring? There are many different questions that can be asked in this area and there is an appropriate situation for every method, so no one answer is correct. However, the ultimate goal is to gain an understanding of how an order is being handled so the process is transparent and understood.

Analysing protections

Finally, there are many different ways a counterparty can protect your order when interacting with the market, therefore which of these, if any, are your counterparties using and in what circumstances.

For example, does your counterparty’s algorithm utilise a Minimum Execution Size (MES) or a Minimum Average Size (MAS) to protect against small fills? Are these nuanced for different venue types or even individual venues? For instance, is the same MES used for periodic auctions vs. conditional orders on dark MTFs? Intuition would say this doesn’t make sense given the very different average execution sizes on each.

Are limit prices placed on child orders sent to the market and how are these set in relation to any parent limit provided? Limit prices offer one of the simplest ways to protect against sudden price movements or changes in the primary order book spread and liquidity. Some counterparties offer the ability to adjust the sensitivity of any child limit price model, so is this set up appropriately for your flow?

A participation rate constraint can also help to protect against sudden surges in volume at a particular price point; is this something utilised by your counterparty and, again, is it set up appropriately

Conclusion

It has long been accepted that analysing execution performance is more of an art than a science, however, with the increased scrutiny and monitoring requirements being imposed on the industry, execution evaluation is a process here to stay. Focusing on key areas to analyse and gaining a real understanding of the details can help you recognise what is important to your order and workflow and to help you to evaluate your counterparties, and, ultimately, helping you to see the wood for the trees.

Ellen Gordon