ETFs and Pairs Trading

Pairs trading is a trading strategy in which two securities that are very similar in most aspects, but differ in one key aspect, are traded against each other in a long-short fashion. By going long in one security and short the other, you eliminate most risks except for the one you wish to exploit. This article discusses the strategy and how it is made easier with exchange-traded funds.

In the stock market there is never a sure thing. Even when you are confident about an investment idea, some unforeseen event could wipe out your profits. Pairs trading attempts to control for outside risk and allows you to focus on just one risk at a time. For example, let's say that you are a fan of  Apple (AAPL) products and you anticipated that the iPad would steal market share from  Amazon.com's (AMZN) Kindle. A pairs trade would require you to buy stock in Apple and sell short stock of Amazon. Over the past two months, Apple stock is up about 10% while Amazon is down 5%. The upside of pairs trading is that it reduces market risks, even though it can increase single-security risk. For example, let's say that you witnessed the success of the iPod and how it changed the way people listen to music. Perhaps back in the summer of 2008 you thought that the new version of the iPhone would continue to steal market share from other smartphones such as  Research in Motion's (RIMM) BlackBerry. If you bought stock in Apple in June 2008 without using a pairs trade to eliminate market risk, you would have lost money for at least a year on Apple stock despite the fact that your investment thesis was correct. This is because Apple stock was down about 20% from June 2008 to June 2009, in part because the market was down about 30%. So, if you had used a pairs trade, buying Apple and shorting Research in Motion, the trade would have made a positive 20% because Research in Motion was down 40%, twice Apple's 20% loss.

Pairs trading has been around for a long time, but how can ETFs make them easier? It is difficult enough to come up with one investment idea, but in pairs trading you need to have two ideas. Buying Apple stock was not enough--you also had to know whom Apple was going to beat and what companies face risk profiles similar to Apple's. On top of this, you had to be able to locate shares of RIMM to short. Because of the variety, depth, and ease of shorting ETFs, you can simply short an ETF within the same asset category, index, or industry as the stock you wish to buy.

Here is an example. Let's say that you fear that the Gulf oil-spill crisis will cause much greater liability and reputational risks to BP than is currently perceived by the market. However, if oil prices go up again, BP stock could go up--but perhaps less than other oil companies. Here, we would short BP stock and go long either an oil industry ETF such as  Energy Select Sector SPDR (XLE) or an oil commodity ETF such as  United States Oil (USO). The oil industry ETF is probably a better choice, because the commodity can move separately from the stock market. This is validated by the fact that BP had a 0.71 correlation to the energy sector ETF but only a 0.60 to USO.

Although correlation is a commonly used statistic to describe the strength of a relationship, a better metric is a stock's beta, which can be used to tell you how many shares you need to short for each long share. For example, let's say you like  Coca-Cola (KO) stock and you want to do a pair trade, going long Coca-Cola and short the  S&P500 SPDR (SPY). These two had a correlation of 0.65 over the past three years. Coke's beta to the market was 0.55, indicating that for every two shares long in Coke, you only need to short about one share of the market. At 0.61,  Monsanto  had a similar correlation to the market as Coke; however, its beta was about 1.0. So, if you were shocked by Monsanto's admission that it would not hit its growth targets and its change in strategy to cut prices on Round-Up, you might want to short the stock. Here, you could purchase SPY at a 1:1 ratio; despite low correlation to the market, Monsanto's greater volatility resulted in a higher beta. Of course, using historical estimates for correlation and beta is risky, as these variables are incredibly unstable.

In summary, pairs trading can be a great way to monetize an investment idea without taking unwanted bets. Because of their liquidity, ease of shorting, and wide number of niche and index product availability, ETFs are the perfect other half of the trade.