Did I get best execution on my institutional order? This question, while seemingly appropriate, doesn’t make sense. It is the wrong question. Best execution is not an outcome - it is a process. The right question is: was my order executed in accordance with our best execution policies and procedures? One way to think about best execution is as due diligence for trade implementation.
Best execution is not an outcome - it is a process.
Why can’t I know if I got the best possible outcome for my order? We can certainly know if we got the best price available at the time. Isn’t this best execution? Unfortunately, no. It is not so simple.
We need to first understand the context of the order. What are the reasons for implementing the trade? Is it an alpha trade? Is it a funding trade? A rebalancing trade? Are we more concerned with cost or risk? Are there any constraints, such as limit prices or volume caps? The answers to these questions will guide our determination of best execution.
Regulators require best execution, but they don’t define it prescriptively. Best execution is instead a principles based rule. This allows for the motivation for the trades to drive the analysis. Given what we are trying to do with our trade implementation, and taking into account best execution, how should we go about ensuring best execution? This is the question both buyside and sellside participants need to consider when executing institutional orders.
Different participants will have different ideas on what constitutes best execution.
Different participants will have different ideas on what constitutes best execution. In addition, the same participant may have different ideas of how apply best execution to different types of orders. Best execution is not a black and white issue, it is shades of grey.
Best execution is data-driven: measurements matter. However, we don't know whether we achieved the best outcome. This is because we can’t roll back time to try other scenarios. Any large order that interacts with the market over a longer time period will need to take into account more than just the best price. The market will respond to our interaction with it making it impossible to know with certainty that our interactions were the best possible interactions. Instead we will need to fall back to statistical analysis and our policies and procedures that define our approach to best execution.
The market will respond to our interaction with it making it impossible to know with certainty that our interactions were the best possible interactions.
What are the components we need to consider? Speed, certainty of execution, overall cost of the order (as opposed to trade-by-trade best price) and how to manage risk are some key aspects. Other important inputs are the reasons behind the order and any instructions and changes to instructions as the order was being implemented.
So, given this situation, how do we use measurements to support best execution? For one, in order to do a full analysis, we will need to record the specific instructions related to the order. We can then tailor our methodology to see if, over a large number of orders, it appears that our best execution policies and procedures are leading to good results. If they are not, we can measure the impact of any changes to our best execution approach that we introduce to try and improve things. In this way we can move toward an optimal approach.
A buyside participant would be interested in measuring the performance of their brokers along the impact cost, venue, routing, broker and algo dimensions. A sellside participant may be interested in measuring liquidity capture of their router, ensuring algos are performing as expected and that client instructions are properly followed. All participants should also be interested in adverse selection (trading too early), opportunity cost (trading too late) and timing of entering the market.
The measurement framework for best execution is transaction cost analysis (TCA).
The measurement framework for best execution is transaction cost analysis (TCA). The TCA framework should allow for more than just cost analysis. Routing, risk and venue analysis are very important in today’s fragmented markets. The data describing the reason behind the trade and any constraints should also be built into the TCA process. While this data may be hard to get, it is invaluable when trying to understand the performance outcomes. TCA also includes pre-trade market impact estimation which can help in alerting traders to unexpected outcomes as trading proceeds.
Best execution is not an outcome, it is a process of applying a series of steps to ensure that orders are implemented in a disciplined manner. The approach should be based on thorough analysis including a data driven feedback loop. Participants can not just purchase best execution solutions off the shelf, they need to create their policies and procedures based on their specific objectives. However, they can leverage measurement frameworks such as TCA to determine, using a data-driven approach, the best way to meet their goals.
To achieve best execution, you need to “get it”.
Comments