The Historical Performance of Equity REITs : a Seasoned Index Approach

The first section of this study tests an investment strategy of buying non-IPO (seasoned) equity REITs as opposed to purchasing equity REIT IPOs. 'Seasoned', as defined in this study, is an interval of time after the REIT IPO date. To test this investment strategy, monthly returns on an equally-weighted index of 66 equity REITs were analyzed. These equity REITs traded on the major stock exchanges for various intervals over 1973-1994. Over the sample period, the total returns of the 'Seasoned Index' were slightly higher than the total returns of the index with no seasoning. More importantly, the seasoning effect was found to be more significant over the second half of the sample period (1984-1994). The findings confirm the hypothesis that there can be slight advantages to investors who wait to buy equity REITs after the IPO date. The second section of this study compares the relative performance of Seasoned Index to that of common stocks, as measured by the S&P 500. Over the entire sample period, the total returns of the Seasoned Index were 37% greater than the total returns of the S&P 500. The empirical section of this study analyzes the risk-adjusted monthly returns of the Seasoned Index to the risk-adjusted monthly returns of the S&P 500. Both a CAPM and five factor Arbitrage Pricing Model found no statistical evidence of excess returns to the Seasoned Index over the S&P 500. Moreover, three of the five macro economic factors seem to consistently drive equity REIT returns: unexpected inflation, the change in the term structure of interest rates, and the change in risk structure of interest rates. The impacts of these variables on equity REIT returns are 68% of the impacts of the returns of the S&P 500. Therefore, returns of the Seasoned Index were found to be less risky than the S&P 500. This aspect of the study is based upon previous research done by Chan, Hendershott, and Sanders(1987). The Seasoned Index in this study included monthly returns on 43 more stocks than the Chan Hendershott, and Sanders data set as well as six additional years of return data and confirms their findings are consistent under present market conditions. Thesis Supervisor: William C. Wheaton Title: Professor of Economics MIT, Department of Economics Thesis Reader: W. Tod McGrath Title: MIT, Center for Real Estate