Academic journal article The Journal of Real Estate Research

Time-Series Properties and Diversification Benefits of REIT Returns

Academic journal article The Journal of Real Estate Research

Time-Series Properties and Diversification Benefits of REIT Returns

Article excerpt

Abstract. This study examines the potential of real estate investment trusts (REITs) to improve the investment opportunity set available to investors in the United States in an ex ante (ie., asset allocation) context. The findings show that conditioning on lagged REIT returns offers investors an improved method to predict volatilities and correlations of REITs with other asset classes. The ex ante benefits of the diversification of REITs are related to ex post performance using a dynamic asset allocation exercise with ex ante information. These portfolios, on average, involve substantial allocation to REITs and achieve mean-variance tradeoffs close to those attained by fixed-weight unconditional mean-variance efficient portfolios.

Introduction

Real estate investment trusts (REITs) have become enormously popular in the past few years. REITs have been around in various forms dating back to the early 1970s and earlier as well. Their sudden rise in popularity has prompted debate on whether the rise in REIT stocks is simply a fad or whether REITs, as an asset class, are here to stay. Authors who have investigated the diversification benefits of real estate investment have found mixed evidence. Many studies have documented that the returns to real estate have very low correlation with movements in the stock market. ' These studies infer that mean-variance optimal portfolios should contain significant allocations to real estate. Many of these studies use appraisal data to compute returns to real estate investments. These data tend to be smoother than the true value series, raising some question as to the true diversification benefits of real estate.

The rise in popularity of REITs has provided researchers with an alternative data source to measure real estate values and to assess the diversification benefits of investment in real estate. The advantage of using REIT market data is that they provide a more up-to-date assessment of the market's valuation of the underlying real estate portfolio. The potential disadvantage is that REITs tend to actively manage their real estate portfolios and hence the market's valuation of REIT stocks reflects not only the value of the underlying real estate portfolio but also investors' expectations regarding the future growth opportunities of REITs. As many researchers have noted, correlations of REIT stocks with the general stock market tend to be much higher than correlations of real estate indices with the stock market. In addition, REITs tend to have low correlation with real estate indices. These findings have led many researchers to suspect the ability of REIT returns to accurately reflect changes in the real estate market. Partly to address these concerns, Giliberto (1993) used market data on REITs and the S&P500 Index to construct a derived series, the hedged REIT Index, that is nearly uncorrelated with the S&P500 Index and more highly correlated with real estate indices, and examines the diversification potential of this index. Giliberto advocates the use of the hedged REIT Index in asset allocation work to avoid the biases in real estate indices due to their use of appraisal data.

Many authors who have investigated the diversification benefits of REITs have found mixed evidence. Burns and Epley (1982) concluded that REITs, in particular equity REITs, offer investors good diversification benefits. Mull and Soenen (1997) concluded that the diversification potential for REITs appears to be very time dependent. They found that while REITs would have been a good investment vehicle in the 1990-1994 period, they were not as attractive in the period 1985-1990. The research cited earlier used the realized covariance matrix of returns to study ex post diversification benefits of REITs. If the realized historical mean return and realized historical covariance matrix over a period of time are good estimates of the future values of these quantities, then these studies may be interpreted in an asset allocation context. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.