Arbitrage has been traditionally defined as the simultaneous buying and selling of a security at two different prices in two different markets, resulting in a riskless profit. As there are no such thing as free lunch in finance, (efficient) markets should not pose any significant opportunity for arbitrage. True arbitrage is hard to find these days. However, multiple arbitrage-like trading strategies have widely researched and deployed. Arbitrage strategies try to identify a situation in which investor profits from identifying mispricing without making any directional bets. Most arbitrage strategies, therefore, try to compose self-financing portfolios that offset each other’s cash flows. This is most commonly achieved by buying and selling similar assets or composing replication portfolios. Profits are viewed as compensation to arbitrageurs for enforcing the “Law of One Price.”

Numerous papers report situations of mispricing which appear to allow for arbitrage profits to be made (e.g. Froot and Dabora, 1999; Mitchell, Pulvino, and Stafford, 2002; and Gagnon and Karolyi, 2010). Others point out that arbitrage is rarely, if ever, as risk-free and costless as the text-book definition suggests (e.g. De Long, Shleifer, Summers, and Waldman, 1990; Shleifer and Vishny, 1997). A basic example is pairs trading strategy with stocks. This is a strategy that was popular on Wall Street during 1990s. A strategy tries to identify two assets that usually trade together in step with a certain long term spread. Once the spread between the assets diverges from normal, an investor buys or sells the spread and waits until the spread returns back to its assumed fundamental value. Gatev et al. (2006) present key research on pairs trading, finding 11% returns on self financial portfolios.

Several studies have made modifications to this pairs trading methodology. Elliott, Van Der Hoek, and Malcolm (2005) use a Gaussian Markov chain model for the spread while Do, Faff, and Hamza (2006) make adaptations for spread measurement based on theoretical asset pricing methods and mean reversion. Thomakos, Wang, Schizas (2011) propose pairs trading strategy on international ETFs. Vidyamurthy (2004) and Burgess (2005) utilize cointegration pairs trading, while Papadakis and Wysocki (2007) expand on the methodology of Gatev et al. (2006) by examining the impact of accounting information events (i.e. earnings announcements and analyst forecasts) on the level of returns of the pairs trading strategy. More recently, Do and Faff (2012) present US evidence on the effect of transaction costs on strategy profitability. For US studies, the strategy reports the highest level of profitability during the 1970s and 1980s, with a notable decline after 1989.

Despite considerable theory about market efficiency, economists have little empirical information about how efficiency is maintained in practice. Arbitrage strategies may seem low risk in most normal market settings, but can also amplify losses during rare market events. The most famous example of an arbitrage gone wrong is a fixed income arbitrage done by Long Term Capital Management LTCM. LTCM deployed sophisticated quantitative strategies executed by computers and designed by some of the top finance academics during the late 1990s. After a very successful few years, LTCM had run into problems when some of their fixed income pairs did not converge, and the company was forced to close their positions due to inability to support the strategy with new funds. In the end, LTCM had to be bailed out for 4 billion USD to prevent systemic effects.

Subscribe for Newsletter

Be first to know, when we publish new content

The Encyclopedia of Quantitative Trading Strategies

Log in