Conditional Volatility of Equity Real Estate Investment Trust Returns: A Pre- and Post-1993 Comparison

Abstract

We examine the dynamic behavior of Equity Real Estate Investment Trust (EREIT) volatility in a GARCH context 1972–2006 using monthly EREIT returns, and comparing volatility performance for “early” Equity REITs 1972–1992 with that of “modern” EREITs 1993–2006. Consistent with findings for conventional firms, we find that EREIT conditional volatility is time-varying, persistent, and predictable. There is a positive relationship between expected return and expected risk in EREIT stocks pre-1993, but the relationship disappears after 1993. We find no evidence that negative shocks affect EREIT volatility differently from positive ones in either time period. Different from reported results for conventional firms, we find that changes in the conditional volatility of fundamental macroeconomic variables have strong explanatory value for future changes in EREIT volatility. Finally, comparing EREIT volatility performance with volatility in the Russell 2000 Index, a proxy for small stocks, we find that EREIT volatility behaves differently from that of small stocks in many respects, indicating that risks in the small stock index cannot effectively proxy for risks in the EREIT market. Copyright Springer Science+Business Media, LLC 2009Real estate, Real estate investment trusts, REITs, GARCH, Price volatility, Conditional volatility, Volatility persistence, Macroeconomic volatility, C14, C22, C32, F31, G12, G15,

Volatilities and Momentum Returns in Real Estate Investment Trusts

“…Black and McMillan (2006) apply a GARCH-in-mean model and show that value stocks have higher asymmetric risk than growth stocks, therefore explaining the value premium (value stocks with high book-to-market ratio have higher returns than growth stocks with low book-to-market ratio). In the real estate sector, Jirasakuldech et al (2009) estimate the conditional volatility of equity REITs in a GARCH model, and compare the conditional volatility for two sub-periods: pre-1993 and post-1993. The authors find a positive relationship between conditional volatility and expected return during the pre-1993 period, but find no evidence of any significant relationship during the post-1993 period.…”

A Comparison of Alternative Forecast Models of REIT Volatility

“…For example, Bollersev (1986) reported that GARCH(1,1) is an appropriate model to characterize the volatility behavior for the general U.S. stock market. In the REIT literature, Stevenson (2002), Cotter and Stevenson (2006), and Jirasakuldech et al (2009) showed that GARCH(1,1) is adequate to model REIT volatility. Cotter and Stevenson (2008) also used FIGARCH(1,1) and FIEGARCH(1,1) to model the persistence in REIT volatility.…”

“…A number of studies have been devoted to the examination of occurrence of structural breaks in the REIT sector. See, for example, Chui et al (2003), Ewing and Payne (2005), Jirasakuldech et al (2009), Kim et al (2007, and Okunev et al (2000). If structural breaks did occur, we could divide the full sample into several subsamples based on the estimated break dates.…”

Idiosyncratic risk and the cross-section of European real estate equity returns

“…Depending on the return frequency used to calculate idiosyncratic risk (daily vs monthly), portfolio or individual stock level data, return weighting procedures, idiosyncratic volatility measures and general model set-up, idiosyncratic risk is found to be either positively, negatively or unrelated to expected stock returns. Concerning the US-REIT market, the studies of Jirasakuldech et al (2009), Chiang et al (2009), Ooi et al (2009), Liow and Addae-Dappah (2010) and DeLisle et al (2011) add to the inconclusive evidence on the pricing of idiosyncratic risk.…”