How Program Trading Strategies Work

Program trading strategies happen behind the scenes, trading up to 30% of the daily volume at the NYSE. They trade without emotion and can be highly profitable. These trades, called program trades, take place quietly, oblivious to the chaos of the trading floor. However, savvy investors would be foolish to ignore how and why a system that produces an average of 30% of the daily trading volume on the New York Stock Exchange (NYSE) does so well.

In general terms, program trading is large-volume trading made by systems, usually automated, based on an underlying program or idea. However, there is more to program trading than this simple definition implies. The NYSE defines program trading as a "wide range of portfolio trading strategies involving the purchase or sale of 15 or more stocks having a total market value of $1 million or more."

The term "system trading" is frequently used interchangeably with program trading; however, this is not entirely accurate. System trading refers to a methodology that may produce program trading if done in sufficient volume. Conversely, certain program trades can be generated by a system-trading methodology. Program trading, for our purposes here, refers only to the NYSE definition.

Contrary to popular belief, the underlying portfolio strategy behind a program buy or sell is often not computer-generated. The goals may be as disparate as portfolio balancing to broad asset allocations to sector allocations. They may be intraday strategies, short-term, or long-term strategies. The actual strategies, and the algorithms that generate program buys and sells are proprietary to each player and are among the most closely guarded secrets on Wall Street.

Program trades are almost always executed by computers, although there are instances when this isn't the case. For example, if institution XYZ wants to sell a basket of 15 stocks totaling $2 million, it could simply split up the sale among several different brokers. Conversely, a big buy program on a single stock may go directly to a market maker or to a single broker who then splits it up into smaller units. As a practical matter, the NYSE is only interested in regulating the computer-generated program trades, and in particular, those generated by large movements in the futures premium.

Importantly, much of program trading involves the futures markets as well as the cash market. The most simplistic and widely known of these strategies is index arbitrage. Index arbitrage is frequently used by institutions with very large and diverse stock portfolios under management.

For example, an institution buys futures when the premium is low, while it simultaneously sells a basket of stocks in a hedged trade to garner a few points of return over what a portfolio of S&P stocks would produce on its own.

The important point for the individual investor is that the futures market and the cash market are intimately intertwined. Moves in one market can trigger moves in the other. Every day, the S&P futures have a fair value based on a formula that includes, for example, days to expiration and the cost of carry for a commensurate basket of stocks.

There are certain levels of premium that will generate program trades, although it varies slightly among firms due to different costs of carry. Every day there are "buy execution levels" and "sell execution levels." The best (and only public) source of information for daily fair value and premium execution levels can be found at HL Camp & Co.'s Program Trading Research site. Additionally, the NYSE publishes program trading activity by member firms every week for the previous week at its website. This is interesting reading, but not particularly useful for real-time decisions.

During the 1980s and '90s, program trading was largely blamed for excessive volatility in the stock market and was named as a culprit in some major crashes. As a result, the NYSE has imposed rules that define certain times when computer-generated program trading is restricted.

Since the new rules were established, there have been very few disruptions directly attributed to program trading. Given the amount of liquidity that program trading contributes to the stock and futures markets, its effect is probably more beneficial than not, even during sharp corrections.

Program buying and selling has a tendency to happen at certain times of the day, sometimes called reversal times. Over time, these will become evident to the attentive observer by spikes in volume and wider price swings. Why is this important? Let's say you own a Dow stock and want to sell it, wouldn't it help you to do so during a buy program?

Program trades represent a large chunk of any given day's market activity and their impact on market movements is important. During the slower summer months, program trades make up as much as 50% of market activity. Smart investors must watch for patterns and time their buying and selling to make sure they aren't caught on the wrong side of these large-volume computer-controlled trades.