Securitised Real Estate Regime-Switching Behaviour and the Relationship with Market Interest Rates

This paper examines the regime-switching behaviour of the six major international securitised real estate markets by utilizing the Dynamic Markov-Switching methodology. Using crises-rich sample period of 1993-2010, we find that international securitised real estate returns can be sufficiently characterised by two distinct domestic regimes of high return-low volatility and low-return high-volatility that are substantially different from the underlying stock markets in terms of the duration and risk-return characteristics of the regimes in each market. The presence of two distinct regimes provides new international evidence on the sector’s relationship with interest rates and the fact that securitised real estate reaction to unexpected market interest rate changes is of the asymmetric nature.We also analyse each countries securitised real estate volatility reaction to the state-dependent measures of the interest rates uncertainty using GARCH models. GARCH models are augmented with conditional volatilities of the interest rate variables and are estimated with the maximum likelihood functions suggested in Bollerslev & Wooldridge (1992). We find that the volatility of securitised real estate markets is affected by the interest rates uncertainty across our sample. Moreover, we find consistently that higher variance in securitised real estate is associated with either the absolute change or increased conditional volatility of domestic interest rates. The asymmetry in the variance sensitivity to interest rates is dominated during regimes shifts within the listed real estate sector and not the overall stock market. In the latter case, the evidence of asymmetries in variance is much weaker. A potentially mistaken assumption that securitised real estate markets are driven by the underlying stock markets explains a lack of significant findings in Chen et al. (2012). One the last two findings reveal a counter-cyclical behaviour of the securitised real estate sector when compared to the stock market. Therefore, the real estate sector yet again can be viewed as a separate market from the general equities. This can be potentially useful from the portfolio management point of view since we find that the impact of the potential risk proxies is weaker during the periods of the general market instability.