The dynamics of return co-movements and volatility spillover effects in Greater China public property markets and international linkages
| Published date | 26 August 2014 |
| Pages | 610-641 |
| DOI | https://doi.org/10.1108/JPIF-06-2014-0039 |
| Date | 26 August 2014 |
| Author | Kim Hiang Liow |
| Subject Matter | Property management & built environment,Real estate & property,Property valuation & finance |
The dynamics of return
co-movements and volatility
spillover effects in Greater
China public property markets
and international linkages
Kim Hiang Liow
Department of Real Estate, National University of Singapore,
Singapore, Singapore
Abstract
Purpose – The purpose of this paper is to examine weekly dynamic conditional correlations (DCC) and
vector autoregressive (VAR)-based volatility spillover effects within the three Greater China (GC) public
property markets, as well as across the GC property markets, three Asian emerging markets and two
developed markets of the USA and Japan over the period from January 1999 through December 2013.
Design/methodology/approach – First, the author employ the DCC methodology proposed by
Engle (2002) to examine the time-varying nature in return co-movements among the public property
markets. Second, the author appeal to the generalized VAR methodology,variance decomposition and
the generalized spillover index of Diebold and Yilmaz (2012) to investigate the volatility spillover
effects across the real estate markets. Finally, the spilloverframewo rk is able to combine with recent
developments in time series econometrics to provide a comprehensive analysis of the dynamic
volatility co-movements regionally and globally.The author also examine whether there are volatility
spillover regimes, as well as explore the relationship between the volatility spillover cycles and the
correlation spillover cycles.
Findings – Results indicate moderate return co-movements and volatility spillover effects within and
across the GC region. Cross-market volatility spillovers are bidirectional with the highest spillovers
occur during the global financial crisis (GFC) period. Comparatively, the Chinese public property
market’s volatility is more exogenous and less influenced by other markets. The volatility spillover
effects are subject to regime switching with two structural breaks detected for the five sub-groups of
markets examined. There is evidence of significant dependence between the volatility spillover cycles
across stock and public real estate, due to the presence of unobserved common shocks.
Research limitations/implications – Because international investors incorporate into their
portfolio allocation not only the long-term price relationship but also the short-term market volatility
interaction and return correlation structure, the results of this study can shed more light on the extent
to which investors can benefit from regional and international diversification in the long run and
short-term within and across the GC securitized property sector, with Asian emerging market and
global developed markets of Japan and USA. Although it is beyond the scope of this paper, it would be
interesting to examine how the two co-movement measures (volatility spillovers and correlation
spillovers) can be combined in optimal covariance forecasting in global investing that includes stock
and public real estate markets.
Originality/value – This is one of very few papers that comprehensively analyze the dynamic return
correlations and conditional volatility spillover effects among the three GC public property markets, as
well as with their selected emerging and developed partners over the last decade and during the GFC
period, which is the main contribution of the study. The specific contribution is to characterize and
measure cross-public real estate market volatility transmission in asset pricing through estimates of
several conditional “volatility spillover” indices. In this case, a volatility spillover index is defined as
share of total return variability in one public real estate market attributable to volatility surprises in
another public real estate market.
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1463-578X.htm
Received June 2014
Accepted June 2014
Journal of Property Investment &
Finance
Vol. 32 No. 6, 2014
pp. 610-641
rEmeraldGroup Publishing Limited
1463-578X
DOI 10.1108/JPIF-06-2014-0039
610
JPIF
32,6
Keywords Error-correction model, Correlation spillover cycle, Directional and net volatility spillovers,
Dynamic conditional correlations, Greater China public property markets, Volatility spillover effects
Paper type Research p aper
1. Background, motivation and research questions
In recent years, portfolio investors and academic researchers have devoted increasing
attention to the issues of public property market linkage and financial contagion
because international public real estate diversification might be more effective than
international stock diversification. As have been documented in the finance literature,
the two most popular measures of co-movement are return cor relations and volatility
spillovers. Although there is considerable academic literature on the extent to which
international public stock (real estate) markets have co-moved with one another based
on correlation coefficient analysis, to our knowledge less formal studies have focussed
on the extent and nature of volatility interdependence and volatility spillovers among
the three Greater China (GC) public real estate markets, Asian emerging markets and
developed markets of Japan and the USA. With international financial markets
becoming more correlated and connected than ever before, a good understanding
regarding the evolution and nature of volatility transmission, the intensityand direction of
volatility spillovers, as well as the correlation variation and spillovers over time provides
useful information regarding market co-movement dynamics for investors, financial
institutions and policy makers. The ongoing process of globalization which was greatly
simulated by market-opening policies in many emerging countries during the late 1980s,
also contributes to an increasing interest in the study on financial market co -movements
including public property markets. With remarkable growth in the securitized real estate
sector over the last two decades, the public property sector is now an “essential” asset
class in domestic/international mixed-asset portfolios. This development reinforces the
specific importance of this topic. Moreover, owing to the rapid growing size of the Chinese
stock markets, as well as the growing presence of the GC economy (comprising Mainland
China, Hong Kong and Taiwan), a study on the co-movement dynamics among the three
GC markets, as well as with other Asian emerging markets/developed markets is of
paramount interest. This study focusses on public real estate markets.
The core objective of this paper is to comprehensively analyze the dynamic return
correlations and conditional volatility spillover effects among the three GC public
property markets, as well as with their selected emerging and developed partners over
the last decade and during the global financial crisis (GFC) period, which is the main
contribution of the study. Our contribution is to characterize and measure cross-public
real estate market volatility transmission in asset pricing through estimates of several
conditional “volatility spillover” indices. In our case, a volatility spillover index is
defined as share of total return variability in one public real estate market attributable
to volatility surprises in another public real estate market.
The second contribution of ou r study is the appli cation of newly generalized
version of the spillover index of Diebold and Yilmaz (2012). This generalized
spillover approach produces variance decompositions which are insensitive to the
variable ordering by allowing correlated shocks and using the historically observed
distribution of the errors to account for the shoc ks, and thus represent a significant
improvement over the traditional Cholesky-factor identification of vector
autoregressive (VAR) (Gaspar, 2012). Moreover, as pointed out by Yilmaz (2009),
since the spillover index is based on a multivariate VAR it can be better utiliz ed to
capture the increased co-movement of business and financial market fluctuations in
611
GC public
property markets
and international
linkages
more than two countries compared to an analysis based on bivariate correlation
coefficients. We employ this method to estimate the total volatility spillover indices,
gross directional spillover indices, net directional spillover indices and rolling spillover
indices of the sample public real estate markets.
Third, the generalized spillover approach treats the rolling (time-varying) spillover
index (including rolling gross directional spillover and rolling net spillover indices) as
a variable with a time series structure/evolution. Consequently, the spillover framework
is able to combine with recent developments in time series econometrics to provide
a comprehensive analysis of the dynamic volatility co-movements regionally and
globally. We examine whetherthere are volatility spillover regimes, as well as explorethe
relationship between the volatility spillover cycles and the correlation spillover cycles.
The final contribution of our study is to evaluate the relationship between the total
volatility spillover index of public real estate and stock markets. Since public real
estate market is an imperative part of stock market in many Asian economies, readers
would be interested to know to what extent the total multivariate volatility spillover
index of the two asset markets are linked. The basic idea underlying the modeling
approach we adopt is popularly known as the “error correction model – ECM.” If the
two volatility spillover indices are co-integrated and are subject to an ECM, than we
might be more inclined to believe that regional and global financial integration would
further reduce the potential gains from diversifying into the stock and public real
estate portfolios. Accordingly, our research questions include the following:
RQ1. What is the current level of conditional correlations among the GC public real
estate markets, as well as between the GC markets and emerging and
developed markets?
RQ2. What is the current level of volatility spillovers within and across the GC
public real estate markets?
RQ3. How much of the volatility spillover effects can be attributed to a specific
market; or to what extent does a specific market transmits (or receives)
spillover effects to (from) other markets?
RQ4. What is the behavior of volatility spillover effects during the GFC period?
RQ5. Are the spillover indices subject to regime changes?
RQ6. Are the volatility spillover and correlation spillover indices synchronized?
RQ7. Are the stock and public real estate volatility spillover effects in long-run
equilibrium?
The remainder of this paper is organized as follows. Section 2 provides a brief review of
the relevant literature. Sections 3 and 4 describe the data and methodology. The
presentation of the empirical results than follows in Section 5, and Section 6 concludes.
2. Brief literature review
Many stock market studies have investigated cross-market short-term movement on
the basis of return correlations and the transmissio n of stock return and volatility
612
JPIF
32,6
To continue reading
Request your trial