Return and co-movement of major public real estate markets during global financial crisis. A frequency domain approach
| DOI | https://doi.org/10.1108/JPIF-01-2017-0002 |
| Pages | 489-508 |
| Date | 07 August 2017 |
| Published date | 07 August 2017 |
| Author | Kim Hiang Liow,Shao Yue Angela |
| Subject Matter | Property management & built environment,Real estate & property,Property valuation & finance |
Return and co-movement of major
public real estate markets during
global financial crisis
A frequency domain approach
Kim Hiang Liow and Shao Yue Angela
Department of Real Estate, National University of Singapore, Singapore
Abstract
Purpose –The purpose of this paper is to investigate the volatility spectral of five major public real estate
markets, namely, the USA, the UK, Japan (JP), Hong Kong (HK), and Singapore (SG), during the pre- and post-
global financial crisis (GFC) periods.
Design/methodology/approach –First, univariate spectral analysis is concerned with discovering price
cycles for therespective real estate markets. Second,bivariate cross-spectralanalysis seeks to uncover whether
any two real estate price series share common cycles with regard to their relative magnitudes and lead-lag
patterns of the cyclical variations. Finally, to test the contagion effects, the authors estimate the exact
percentagechange in co-spectral density(cyclical covariance) dueto high frequencies (short run) afterthe GFC.
Findings –The authors find that whilst none of the public real estate markets examined are spared from the
crisis, the three Asian markets were less severely affected by the GFC and were accompanied by a reversal in
volatility increase three years post-global financial crisis. Additionally, the public real estate markets
studied have become more cyclically linked in recent years. This is particularly true at longer frequencies.
Finally, these increased cyclical co-movements measure the outcomes of contagion and indicate fairly strong
contagious effects between the public real estate markets examined due to the crisis.
Research limitations/implications –The implication of this research is that benefits to investors from
international real estate diversification may not be as great during the present time compared to previous
periods because national public real estate markets have become more correlated. Nevertheless, the findings
do not imply the complete absence of diversification benefits. This is because although cyclical correlations
increase in the short run, many of the correlation values are still between low and moderate range, indicating
that some diversification benefits may still be realized.
Practical implications –Given the significant market share and the highest levels of securitization in
Asia-Pacific markets including JP, HK/China, and SG, this cyclical research including major public real estate
markets has practical implications for ongoing international real estate investment strategies, particularly for
the USA/UK and Asian portfolio managers.
Originality/value –This paper contributes to the limited research on the cyclical return and co-movement
dynamics among major public real estate markets during financial/economic crisis in international finance.
Moreover, the frequency-domain analysis conducted in this paper adds to better understanding regarding the
impact of GFC on the cyclical return volatility and co-movement dynamics of major developed public real
estate markets in international investing.
Keywords Global financialcrisis, Contagion, Frequency domain, Co-movement, Public real estatemarkets,
Volatility spectra
Paper type Research paper
1. Introduction
In this paper, we study the return and co-movement dynamics of five major public
(securitized) real estate markets, namely the USA, the UK, Japan ( JP), Hong Kong (HK), and
Singapore (SG) during the pre- and post-global financial crisis (GFC) periods. In contrast to
the time domain approach, we adopt the frequency domain analysis to show at which
frequencies the returns and co-movements are active. Given the context of economic
globalization and frequent occurrence of financial crises over the past years, the GFC
provides an excellent empirical opportunity to study the changes of the cyclical components
in asset market returns and volatilities, thereby providing fresh implications for market
interdependence and diversification. A related issue is whether any observed higher market
Journal of Property Investment &
Finance
Vol. 35 No. 5, 2017
pp. 489-508
© Emerald PublishingLimited
1463-578X
DOI 10.1108/JPIF-01-2017-0002
Received 1 January 2017
Revised 19 April 2017
Accepted 22 April 2017
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1463-578X.htm
489
Public real
estate markets
co-movements can be attributed to integration or contagion. This paper contributes to the
limited research on the cyclical return and co-movement dynamics among major public real
estate markets during financial/economic crisis in international finance.
In this study, “real estate market”refers to securitized real estate market or public real
estate market. We select public real estate markets in this study because international real
estate diversification might be more effective than international stock diversification
(Eichholtz, 1996). Real estate markets are an important component of the global stock
market, and are always affected before other industries by major events and regulations.
Previous research studies have acknowledged the importance of real estate as an
“alternative”investment asset to obtain diversification benefits in a mixed-asset portfolio.
Moreover, with remarkable growth in the property industry due to real estate asset
securitization and the establishment of real estate investment trust (REIT) vehicle in many
countries over the last two decades, the real estate sector has now been recognized as an
essential asset class in investors’mixed-asset portfolios, in addition to its existing role as an
“industry”of the domestic stock market (Dhar and Goetzmann, 2006). Thus, the inclusion of
real estate in this study reflects the increasing important role of this new “asset”class in
international financial markets.
Spectral/co-spectral analysis is the proper econometric tool to deliver economic insight
for this research. Specifically, we focus on an examination of the return spectra over
different periods of time. We also examine the cross-spectra of the five real estate markets in
the pre- and post-GFC periods with regard to their changes in the cyclical relationship.
To our knowledge, although correlation analysis is the most basic and popular approach to
study the degree of co-movement between economic variables and between financial
markets, it is basically a static analysis which fails to capture the dynamics in the market
co-movement adequately. In contrast, frequency domain approach is particularly useful in
detecting cycles, common cycles, and their linkages. Additionally, its co-spectral analysis
generates the power spectrum to the two series case and provides an advanced method for
interpreting the relationship between a pair of series, as well as determining the relative
importance of cycles of different frequencies in accounting for the market volatility
co-movement of the five economies during the two periods examined. For this purpose, we
employ a unique weekly data set that includes the USA and other four major real estate
markets over the period from July 14, 2000 to June 29, 2007 (pre-crisis period) and from
July 6, 2007 to June 27, 2014 (crisis/post-crisis period). A longer view (seven-year period) is
taken to compare the return spectra and to evaluate the pre-and post-GFC cyclical returns
and co-movements from four perspectives: power spectrum/ co-spectrum, coherence, phase
lead, and variance-covariance. Taking the results together they may provide additional
implications with respect to information transmission among markets, market
interdependence, and diversification.
Given the limited existing empirical work on the cyclical returns and their co-movements
around the GFC at the international and regional levels, the results from this study should
be interesting and useful for global/Asian investors, policy makers, and empirical
researchers. The plan of this paper is as follows. Section 2 provides a brief review of the
literature. This is followed by an outline of the spectral and co-spectral method in Section 3,
an explanation of the data requirement in Section 4 and a discussion of the empirical results
in Section 5. Section 6 concludes the study.
2. Previous literature
Co-movements of stock markets have become a hot research topic in the literature
(e.g. Hamao et al., 1990; Janakiraman and Lamba, 1998; Cha and Oh, 2000; Tay and
Zhu, 2000; Chan et al., 2008). These studies employ the usual time-domain approach such as
time-series linear or non-linear co-integration, vector error-correction, vector autoregressive,
490
JPIF
35,5
To continue reading
Request your trial