“Happiness and freedom begin with a clear understanding of one principle: Some things are in our control, and some things are not. It is only after you have faced up to this fundamental rule and learned to distinguish between what you can and can’t control that inner tranquility and outer effectiveness become possible.”
One of the central tenets of Stoicisim, an ancient Greek philosophy, is that there are things in our lives that we can control and there are things that we cannot control. From the perspective of best execution and transaction cost analysis (TCA), we can identify both the things that we have control over and the things that we don’t have control over when implementing an order in the market.
The goal for a trader is to be able to optimally interact in the market while managing the many dimensions involved in the process. The challenge faced by traders is to interact with the market to achieve “best execution” of the outcomes. While the trader does not have control over things like market liquidity and volatility etc., the trader’s interactions with the market can certainly affect the liquidity, volatility and other factors beyond the trader’s control. This is really the point of the market, to incorporate all the various interactions of many traders into the price discovery process.
Traders usually do have the ability to control how they interact with the market. They have an intuitive understanding that different tactics should be used in different market conditions and for securities that have different liquidity profiles. To this end, they can determine things like strategy selection and the pace of execution as well as the broker they consider to be in the best position to execute the order.
Things traders control:
- Timing of the order
- Picking a broker (buy-side trader)
- Choosing an implementation strategy
- Pace of the execution strategy
Things traders don’t have control over:
- Behaviour of other market participants
- Explicit instructions
When it comes to TCA, we can group performance factors into those that we have control over and those that we don’t have control over. We can then analyse the results to determine the best way to utilise the factors we have control over when we place our next order given a set of circumstances that we don’t have control over. For example, given a market condition, we can choose the appropriate strategy and pace of the order we want to implement. We can also use these factors, along with an analysis of past results, in predictive modelling of things like expected market impact.
TCA provides the framework for these measurements. The onus is on the user to apply weightings to the various factors because each firms’ goals may be specific to that firm. There may also be different approaches within a firm based on the investment and trading goals of specific internal groups. These internal groups may each want to apply different weightings to the various factors that impact performance based on their specific goals. From a TCA perspective, this can be achieved using additional attributes to allow partitioning of the orders into groups with differing utility functions. By marking the orders with identifiers, different weighting schemes can be applied when measuring the post-trade results of the implementation of the order in the market.
One way to think about this is that traders need to operate in a market environment that is itself, subject to change. The trader can’t control the overall market conditions – these conditions result from the aggregate actions of all traders – but the important point to make is that they are able to control how they respond by taking actions that are within their control. They can measure the results and then analyse and learn from those results through usage of a TCA framework which can lead to better future decisions.
By recognising that certain factors are beyond the control of the trader, the TCA results can be used in an attribution analysis to determine how much of an order’s slippage is due to factors beyond a trader’s control and how much is due to factors within a trader’s control. By doing this, the trader can focus on optimising the controllable factors given a set of uncontrollable factors. The trader can attribute the performance due to uncontrollable factors versus the performance due to the trader’s actions. As a simple example, it is a possible to compare a securities’ intraday move in tandem with the move of a major index which the stock is associated with. The price change can then be “beta” adjusted to provide a probabilistic estimate of how much the index move affected the security. Similar approaches for other liquidity analytics are possible and can help to identify the best places for a trader to focus on improving outcomes. There is no point trying to control the uncontrollable, while at the same time there is every reason to mange the controllable decisions.
The result, if Epictetus was right, should be happier traders generating more efficient outcomes.
Written by Chris Sparrow and Henry Yegerman at LiquidMetrix