The popular comic book character Batman was well-known for the great diversity of tools at his disposal when battling crime in Gotham City. As a superhero, he was unique in that he had no supernatural powers - his success was completely driven by the inventive and aggressive adaptation of new tools (all annoyingly named to begin with the word "bat").
Mercer Oliver Wyman (New York) believes that the buy-side trading community could learn from his example. Buy-side traders now have a rich set of algorithmic trading capabilities on their "utility belts," and they need to employ them more frequently and in more-creative ways in order to grapple with the ongoing pressures to reduce transaction costs, increase trade anonymity and, most important, free up critical time to work complex orders that demand their expertise and experience. But there are questions to be asked and information to consider before jumping in.
What are the roadblocks you need to clear?
Three issues are primarily responsible for the lack of algorithmic trading diversity and creativity.
1. Limited view of strategic options.
Algorithmic trading presents a number of tough strategic questions that, left unanswered, will impede the growth and creative use of these new tools:
- What is the impact of algorithmic trading on my trading strategies and costs?
- How do I develop a framework for evaluating algorithmic trading versus normal execution services from my brokers?
- How do I weigh the need for research, market color and other "soft" services that I may not receive using algorithmic trading?
- What internal training, sell-side assistance and incremental resources are required to master these new skills?
- How can I extend algorithmic trading lessons learned to other trading desks besides cash equities (e.g., fixed income, foreign exchange, exchange-traded derivatives)?
2. Too many options to consider.
Buy-side firms face a daunting set of choices when approaching the decision to expand their use of algorithmic trading. The first difficult decision is the choice of the algorithms themselves - most providers offer at least 10 different algorithms as part of their basic algorithmic service offerings. Firms must evaluate algorithms against a structured set of measurements (e.g., speed, transaction cost impact, anonymity, flexibility, complexity, failure rate, usage costs). This is a critical first step, which many firms ignore and, thus, begin down a path that may not be optimal for given execution strategies. The following chart offers just a few examples of standard algorithms:
Unfortunately, the choice of algorithms is not the most complicated decision facing buy-side firms - the way to implement them is. There are a host of providers offering solutions, so the algorithm market offering is diverse.
Many firms may have gained exposure to algorithmic trading through their traditional broker-dealer contacts, but there are other viable options for taking advantage of these capabilities with smaller agency brokers or in the customization of vendor tool kits. The following are the major categories of algorithmic and technology providers:
3. Technology issues.
Once buy-side firms weigh the variety of vendor options, business and IT leadership need to consider:
- How can multiple algorithmic trading systems be cleanly integrated with current order management systems to ensure that they don't overload buy-side traders?
- What is the most effective route for integrating execution systems with internal pre-trade analytics, third-party transaction cost analysis tools and sell-side analytics?
- How should we leverage the enhanced execution speed now available by colocating algorithmic trading equipment at relevant exchanges and trading hubs?