Date Stamping Bubbles In Real Estate Investment Trusts (REIT)
The University of Texas Rio Grande Valley; Texas A&M University – Department of Economics
The University of Texas Rio Grande Valley
October 19, 2015
We test for the existence of single and multiple bubble periods in four Real Estate Investment Trust (REIT) indices using the Supremum Augmented Dickey-Fuller (SADF) and the Generalized SADF. These methods allow us to estimate the beginning and the end of bubble periods. Our results provide statistically significant evidence of speculative bubbles in the REIT index and its three components: Equity, Mortgage and Hybrid REITs. These results may be valuable for real estate financial managers and for investors in REITs.
Date Stamping Bubbles In Real Estate Investment Trusts (REIT) – Introduction
Academics have suggested and employed various time series methods to capture speculative bubbles in asset prices such as integration and cointegration tests (Diba and Grossman, 1988a and 1998b), variance bound tests (LeRoy and Porter, 1981 and Shiller, 1981), specification tests (West, 1987) as well as Chow and CUSUM-type tests (Homm and Breitung, 2012). The new recursive flexible window right-tailed ADF testing procedure introduced in Phillips, Wu and Yu (2011) and further enhanced in Phillips, Shi and Yu (2015) outperforms preceding methods in detecting and date-stamping bubbles and can serve as a real-time warning signal to monitor the dynamics of asset prices.
In this paper, we employ the Phillips, Shi and Yu (2015)’s novel Generalized Supremum Augmented Dickey-Fuller (GSADF) to test for the existence of speculative bubbles and to identify the origination and the collapse of bubbles in various Real Estate Investment Trust (REIT) indices.1 Specifically, we search for explosive autoregressive behavior in inflationadjusted REIT indices from January 1980 through September 2013. We also explain the conditions under which empirical evidence of explosive behavior can be interpreted as a bubble in the price of the underlying financial asset.
The literature on testing for speculative bubbles in REITs is limited and the results are mixed. Jirasakuldech, Campbell and Knight (2006) use unit root and co-integration tests to find that Equity REITs are not affected by rational bubbles. Waters and Payne (2007) use the Residuals-Augmented Dickey-Fuller (RADF) and find no periodically collapsing bubble in total REIT index and Equity REIT index, negative periodically collapsing bubble in Mortgage REIT index and inconclusive results for Hybrid REIT index. Moreover, Payne and Waters (2007) use both Momentum Threshold Autoregressive (MTAR) and RADF to find mixed results for Equity REIT index. Anderson, Brooks and Tsolacos (2011) use regime switching processes (Evans, 1991; van Norden and Schaller, 1999) to directly test for the presence of speculative bubbles in REITs. Although they find some evidence of negative bubbles (most notably in mortgage REIT index), the authors could not observe speculative bubbles in Equity, Mortgage and Hybrid REITs.
There exists important work on the link between REITs, stocks, and real estate markets as well as on speculative bubbles on real estate prices. Goodman and Thibodeau (2008) aim at disentangling the roles of economic fundamentals and speculation on the high house appreciation rates during 2000-2005, while Mikhed and Zem?ík (2009) detect bubbles using panel data on price-rent ratios for the 1975-2006 period and Escobari, Damianov and Bello (2015) propose a time series test to identify housing bubbles. Moreover, Himmelberg, Mayer and Sinai (2005) explain how to assess whether there is a bubble and what underlying factors support housing demand, while Damianov and Escobari (2015) examine the dynamics of price segments during the housing bubble. In a related study, Hendershott, Hendershott and Ward (2003) summarize some evidence on price movements to present arguments for and against the existence of irrational bubbles.
See full PDF below.