The dynamics of risk premium: the case of the Taiwan real estate market

Purpose - This paper aims to examine the risk premium for investors in a changing information environment in the Taiwan, New York and London real estate markets from March 2006 to November 2014. This study attempts to quantify behavioral expectations regarding (or motivation for) investment in the Taiwanese real estate in a changing information environment. Design/methodology/approach - This paper uses the rolling generalised autoregressive conditionally heteroskedastic in mean (GARCH-M) methodology which fixes the problem of conventional GARCH-M methodology. Findings - Empirical evidence suggests that the time-varying risk premium changed for the Taiwan real estate market with a new information set. The risk premium changed from 1.305 per cent per month to -7.232 per cent per month. The study also found persistent volatility shocks from March 2006 to November 2014. No such evidence was found for the New York and London real estate markets. Overall, this study finds evidence of a time-varying risk premium, partly explainable by governmental policies and partly unexplainable. Research limitations/implications - The use of the index of Standard and Poor's Taiwan Real Estate Investment Trusts to study the Taiwan real estate industry may have aggregation effects in result. Practical implications - The present study will provide guidance to investors as well as policymakers regarding the Taiwan real estate market. Originality/value - This study uses the rolling GARCH-M model, which is a first for the Taiwan real estate market.