2011 V39 3: pp. 409–428 DOI: 10.1111/j.1540-6229.2010.00302.x REAL ESTATE ECONOMICS Performance of Pairs Trading Strategy in the U.S. REIT Market Masaki Mori ∗ and Alan J. Ziobrowski ∗∗ We examine the performance of pairs trading in the U.S. REIT market compared with that in the U.S. general stock market over the period 1987 to 2008. The results suggest that the REIT market provided superior profit opportunities for this strategy over common stocks after accounting for the effect of the bid- ask bounce between 1993 and 2000. This was likely because of the unique characteristics of REITs, which permitted the selection of good pairs of close substitutes and the structural changes that occurred in 1993 in the REIT market. The superior trading profits in REITs disappear after 2000. Pairs trading has been one of the most popular quantitative arbitrage strategies used by hedge fund managers on Wall Street. First implemented by Nunzio Tartaglia’s quant group at Morgan Stanley in the mid-1980s (Vidyamurthy 2004), the basic idea behind pairs trading is to find two stocks whose prices have historically moved together and, when the spread between them widens, short the winner and buy the loser. The expectation is that the prices of the two stocks will again converge and the investor will profit by unwinding the positions. Several academic studies propose frameworks to implement pairs trading rather than provide empirical evidence of the effectiveness of pairs trading. Vidyamurthy (2004) explains the link between pricing theory and pairs trading providing an implementation strategy based on cointegration. Elliott, van der Hoek and Malcolm (2005) propose an analytical framework for pairs trading applying a mean-reverting Gaussian Markov chain model. Huck (2009) applies multicriteria decision techniques for the selection of pairs for pairs trading. Regardless of its popularity on Wall Street, only a few studies empirically examine the effectiveness of pairs trading mainly due to the pro- prietary nature of the strategy. Recently, Gatev, Goetzmann and Rouwenhorst (2006) found that using daily historical price information, pairs trading pro- duced significant profits between 1962 and 2002 in the U.S. stock market, which were robust to conservative estimates of transaction costs and were different from profits achievable from a typical contrarian strategy of buying losers and ∗ International University of Japan, Niigata 949-7277, Japan or mori@iuj.ac.jp. ∗∗ Department of Real Estate, Georgia State University, Atlanta, GA 30302-4020 or aziobrowski@gsu.edu. C  2011 American Real Estate and Urban Economics Association