Algorithmic Trading Strategies (Reading 39)

Re: Exhibit 12 on page 48

ABC has low trading urgency while GHIJ has high trading urgency. The text suggests that ABC is best suited for using the VWAP trading algorithm whie GHIJ should use implementation shortfall.

Can anyone get their mind around why ABC should use VWAP instead of implementation shortfall?

VWAP would result in lower commissions ( brokerage costs ) because there is less urgency and can be automated with lower risk of departing from a benchmark. ( however traders could game the benchmark for illiquid or high volume trades particularly their own large orders relative to market , but this is irrelevant ,off-the-mark and unnecessary to add in an essay type question ) Implementation shortfall must be used when trading urgency is high because there is less chance of alpha eroding (particularly if there is a strong trend in the stock )

The background text seems to suggest that I/S is a more all-encompassing way to evaluate a trading strategy, not only because it cannot be gamed, but also because there are 3-4 different factors that measure it’s cost. I suppose I’m struggling with the text, but where does it suggest that VWAP should result in a lower overall execution cost for a low-urgency trade?

I think that would be obvious . Pushing large trades or illiquid trades with urgency implies paying higher costs to brokers. If high urgency trade had the same cost as a low urgency trade , what would the incentive be for the broker to place that trade ahead of others?

VWAP or TWAP Simple Logical Participation (SLP) strategies trade throughout the day to meet their respective benchmarks and minimize trade costs, but fail to consider implicit costs the way Implementation Shortfall does. Implementation Shortfall considers implicit and explicit costs, is not subject to gaming by traders the way VWAP is, is more informative when a trade dominates the daily volume (VWAP will equal that trade so is meaningless), and trades early in the day to minimize delay/missed opportunity trade costs so therefore is better when a trade has high urgency.

Agree with Mark: I would think if ABC is not urgent then a trader could set the trade at VWAP and thereby guarantee that they would at least get the average price for the day. Since a traders’ performance is generally measured against VWAP they wouldn’t take the risk of trading just to minimze the implimentation shortfall, but rather to minimize the risk of underperforming VWAP.