Today’s top firms know there’s more to alpha than just higher trading profits. There’s also minimizing trading costs when implementing your investment strategies. Aside from fixed costs like brokerage and exchange fees, the somewhat hidden costs of execution slippage can erode your alpha and impact overall performance. Not being able to get filled at the price or time you desire creates meaningful consequences, especially when those numbers are compiled over the course of many trades and a significant period of time. Traders must balance price and market impact costs when implementing their investment strategies.
In our technology driven world, many have turned to algorithmic execution to manage and improve their execution costs. And while conversations with experienced users of execution algorithms tend to center around performance and strategy behavior, the pandemic has introduced many discussions with newer users around the basic principles of implementation, setting execution objectives, identifying the appropriate algo classification type (participation, opportunistic or liquidity) and ultimately the best execution strategy for them. In this post we highlight the importance of defining a trader’s execution objective and understanding how different types of execution algos can impact performance.
Imagine trying to choose the best car to get some place fast, but not taking into account the climate or terrain conditions you’ll be driving in, the quantity of passengers you’ll be carrying, or the fuel costs, etc. Without accurate planning, forecasting, and ultimately understanding of the terrain ahead, you may end up selecting a vehicle not suited for your journey. In the end, the vehicle you choose must depend on whether you will often drive alone on smooth roads and in nice weather versus transporting a family in snowy rough terrain. In a similar fashion, there are specific classification types of execution algorithms best suited for different execution objectives.
Once a client has defined their execution objectives, we’ve found that rather than spending too much time reviewing algorithms that don’t meet their unique needs, its best to narrow the focus down to the algo strategies that fall within a specific algo class. There are three popular classifications one can use to classify algorithmic execution strategies (Participation based, opportunistic, and liquidity seeking):
Participation Based Algos: These algorithms are typically reactive in nature and are designed to either participate alongside the current volume or over a certain time frame in the market. Their goal is to blend in and are often used to quietly build large long-term positions or for sizable roll trades. These algos are sometimes also referred to as schedule based algos as a couple of the most commonly used participation based algos like TWAP and VWAP follow defined execution schedules.
Opportunistic Algos: These algorithms attempt to trade in an opportunistic fashion, often seeking to minimize execution costs and/or market impact versus the arrival price (AP) price when the order starts working. Common opportunistic algo strategies are implementation shortfall (IS)/Arrival Price (AP) or passive float strategies.
Liquidity Seeking Algos: Liquidity based algos are designed to find liquidity and transfer risk, typically in short period of time. The popularity of these strategies first grew in the equity markets during the 2008 credit crisis, where time frames were often narrow and being able to find pockets of liquidity was pivotal. With liquidity seeking algos, the user is often of the mindset that the risks of time and price of the market substantially outweigh the concerns over market impact from trading too quickly. Algos in this category have sometimes been referred to as smart market orders, where the user’s primary goal is often order completion in a very short timeframe. Common features often include the ability to sweep multiple levels using IOC order types before temporarily transitioning to a passive stance while the algo’s allow for liquidity to replenish before sweeping again.
Before getting lost in the weeds in trying to understand each underlying algo execution strategy, its best to accurately define your execution objectives and focus on the algo classifications.
Please reach out if we can help define your execution objectives, performance metrics or are interested in learning more about our algorithmic execution types.