Advances in technology and regulation-driven changes in market structure have transformed buy-side trading, most notably in the kinds of trading options available and the strategies needed to ensure best execution for different kinds of trades. Algorithmic trading is one of the fastest-growing of the new technologies, as it has significantly expanded buy-side control over execution and has the potential for lowering transaction costs.

The latest advance in electronic tools, algorithmic trading essentially allows traders to predefine rules for when and how an order should be executed. But it is by no means an autopilot method. While “set it and forget it” approaches may work for infomercials, algorithmic trading requires careful real-time performance monitoring as well as pre- and post-trade analysis to ensure it is properly applied. Traders must calibrate the algorithms to suit portfolio strategy. Far from the sole or final answer to best execution, algorithmic trading represents an additional tool in a trader's expanding kit. The following essay examines best practices in algorithmic trading and how SSgA incorporates them into its trading operations.

Expanded Trading Choices
In the past there were limited opportunities to apply technology to the trading process or interact directly with exchanges and market participants. The trader's main tool was a simple telephone. That all changed with a proliferation of program, direct market access and algorithmic trading as well as rapid growth in electronic markets.

Changes in market structure also advanced the adoption of new technologies such as algorithmic trading. With the introduction of decimalization in 2001, electronic exchanges were able to display quotations in sub-pennies. This has led to some market participants “stepping ahead” of customer limit orders by a small amount to gain execution priority. As a result, many investors have stopped displaying limit orders, reducing market liquidity. Traders have sought more granular methods to break up larger trades into smaller increments. Algorithmic trading has lent itself well to that task.

At the moment there are many execution choices available, which can be categorized on a continuum of “high touch” (i.e., greater human intervention and complexity) versus “low touch” (more automated, less complex) methods. These include (from high to low) single-stock trading, portfolio trading, algorithmic trading, and direct market access (DMA). Each option has its benefits and drawbacks, depending on the nature of a particular trade.

Accessing Liquidity
Amid the increased choices, a trader's core challenge remains efficiently accessing market liquidity. That has become more difficult with structural changes such as decimalization, which further reduced displayed liquidity. Only a small fraction of market liquidity is displayed, such as limit orders from the New York Stock Exchange (NYSE) and electronic limit order books from electronic communication networks or ECNs. A much larger proportion of liquidity is hidden. This encompasses a variety of market, broker-dealer and investor sources including NYSE floor brokers; ECN “reserve” orders; agency orders held by program or cash trading desks and other non-displayed dealer liquidity; as well as orders at buy-side trading desks. A trader's task is to determine how best to deploy the tools at his disposal to access that non-displayed liquidity.

Choosing Wisely
As the number of execution venues has expanded, it is more important than ever for traders to choose wisely for both performance and cost reasons. While superficially algorithmic trading would seem ideal because of its low commissions (in common with other “low touch” tools), it is important to note that commissions represent only a small proportion of total trading costs. According to Wayne Wagner, chairman and co-founder of the money manager advisory Plexus Group, indirect costs due to trading delays, misses or other trading impact issues represent nearly 90% of total trading costs.¹ Any savings from lower commissions can easily be wiped out by an increase in indirect trading costs such as market impact or the opportunity cost of delayed or missed execution. Shifting the wrong kind of trade from a high to low touch venue may reduce commissions, but could result in far higher indirect costs.

Far more important is aligning execution choices with the level of order difficulty involved in terms of: order size, liquidity and trade urgency. Low touch venues such as algorithmic trading lend themselves best to easier types of orders such as low-urgency, small orders for large cap stocks. At the other extreme, urgent orders for a large volume of small cap stocks would require a higher-touch approach to ensure best execution and cost efficiency.

Best Practices at SSgA
SSgA's global equity trading network is a 24-hour operation based in Boston, London and Hong Kong. Processing more than 4,000 trades a day, SSgA traders transacted nearly $200 billion in equity trades in 2004. They are united in their goal of seeking all sources of liquidity to minimize transaction costs. Together with portfolio managers, they are a key factor in successfully implementing investment strategies and growing our clients' wealth.

As algorithmic trading has developed, SSgA traders have worked closely with system providers to fine tune the algorithms used for specific investment strategies. In addition traders draw on the expertise of financial engineers within SSgA's Advanced Research Center (ARC) and our proprietary transaction cost services group, who rigorously evaluate the algorithms and determine where they would be best and most cost-efficiently applied in the trading process. Leveraging SSgA's scale of experience and resources, traders conduct in-depth pre- and post-trade analysis to calibrate algorithms and determine their optimal role in best execution. SSgA maintains an archive of more than 8 million analyzed transactions to guide decision-making. This level of detail helps traders determine which conditions are best suited to algorithmic trading in order to achieve an appropriate balance between direct and indirect costs.

That kind of ongoing research is key to applying the appropriate execution tools for specific strategies amid changing market environments and an increasing flow of electronic trading data. We believe an optimal approach to algorithmic trading requires several critical ingredients:

• Robust pre-trade modeling
• Clear understanding of portfolio management strategies' objectives
• Balancing timing and impact issues
• Intelligently integrating order management systems and direct market access trading platforms
• Close, iterative relationships with algorithmic trading providers
• Thorough post-trade analysis and feedback

Figure 1 summarizes some of the issues involved in choosing the appropriate execution method:

Investment Judgment
SSgA has invested considerable resources into customizing and optimizing algorithmic trading for its trading operations. We recognize that not all algorithms are created equal, and no one solution fits all needs. We also understand that successful implementation requires a deep understanding of the underlying variables as well as real-time performance monitoring and post-trade analysis, as markets continue to change.

While electronic venues are sure to grow in number and use, they will never be a substitute for the investment overview and judgment of an experienced trader. SSgA has developed a very structured process to ensure that the most promising advances in technology are intelligently incorporated into our trading operations. Our commitment to best practices is central to the confidence with which clients outsource their trading activities to us. Our attention to detail, real-time monitoring and continuing research and analysis all support our dedication to better execution and improved portfolio performance for our clients.

¹ Wayne Wagner, testimony to US House of Representatives Financial Services Committee, March 12, 2003.

This material is for your private information. The views expressed are the views of Chris Rice only through the period ended August 9, 2005 and are subject to change based on market and other conditions. The opinions expressed may differ from those with different investment philosophies. The information we provide does not constitute investment advice and it should not be relied on as such. It should not be considered a solicitation to buy or an offer to sell a security. It does not take into account any investor's particular investment objectives, strategies, tax status or investment horizon. We encourage you to consult your tax or financial advisor. All material has been obtained from sources believed to be reliable, but its accuracy is not guaranteed. There is no representation nor warranty as to the current accuracy of, nor liability for, decisions based on such information. Past performance is no guarantee of future results.

Posted On: August 16, 2005