This paper examines the dynamics of return and volatility spillovers across the REIT markets of Japan, Singapore, Hong Kong, Malaysia, Taiwan, Thailand, and South Korea from June 2006 to May 2011. The emerging markets offer lower returns than the developed markets but lower risk as well. The emerging REIT index outperformed the developed REIT index on a risk-adjusted basis. The analysis suggests that correlations among Asian REIT markets are relatively low, ranging from 0.14 to 0.42 over the full-sample period. The results further indicate that correlations among emerging REIT markets are lower than that among developed markets. However, correlations are non-constant over time and increased during the recent Global Financial Crisis. The results from the EGARCH models show that there is a strong tendency for REIT returns to transmit from developed markets (e.g., Japan and Singapore) to emerging REIT markets. In regard to volatility transmission, the mechanism appears to be multidirectional.
Since their establishment in Japan in 2001, real estate investment trusts (REITs) have flourished rapidly across Asia. Asian REITs have become attractive investment vehicles for international funds seeking portfolio diversification (Newell, Liow, Ooi, and Zhu, 2005; Ooi, Newell, and Sing, 2006). As of July 2011, there were more than 95 listed REITs across Asia with a total market capitalization of U.S. $98.7 billion, accounting for 12% of the global REIT market and 5.1% of the global listed real estate securities market (Macquarie Equities Research, 2011). Apart from the success of REITs in the developed markets of Japan, Hong Kong, and Singapore, the last decade has also witnessed impressive growth of the emerging Asian REIT markets, which include Malaysia, Taiwan, Thailand, and South Korea, underpinned by impressive property developments and favorable demographic terms. Despite their growth prospects and increasing significance, investors still have little information and knowledge on these emerging Asian markets. Therefore, a comprehensive understanding of the interrelationships between the emerging and developed Asian REIT markets is crucial for the development of portfolio diversification, trading strategies, hedging, and risk management for investors.
The body of literature on REIT markets in Asia has been thin and modest as compared to that on the United States and Australia. Studies on the performance of Asian REITs in investment portfolios and their correlations with other financial assets had been carried out within a regional context (Chiang, So, and Tang, 2008; Quek and Ong, 2008; Cheok, Sing, and Tsai, 2011; Ong, Ooi, and Kawaguichi, 2011) or for specific countries such as Japan (Kutsuna, Dimovski, and Brooks, 2008; Su, Huang, and Pai, 2010), Singapore (Sing and Ling, 2003), Hong Kong (Newell, Wu, Chau, and Wong, 2010), Malaysia (Newell, Ting, and Acheampong, 2002: Lee and Ting, 2009; Newell and Osmadi, 2009, 2010; Hwa and Wai, 2011), Taiwan (Lin, 2007; Lee, Kuo, Lee, and Lin, 2011), Thailand (Pham, 2011a), and South Korea (Pham, 2011b). Other studies have investigated the linkages in international real estate securities, including the Asia Pacific region (Garvey, Santry, and Stevenson, 2001; Liow, Ho, Ibrahim, and Chen, 2009). These studies, however, concentrate on the listed property companies rather than REITs. The only published paper that examined the volatility transmission of REIT returns in Asia is that of Li and Yung (2007), who investigated the interactions between the Pacific (Australia, Hong Kong, Japan, and Singapore) and the Atlantic REIT markets (U.K. and U.S.). In general, the findings from previous studies can be summarized as: (1) Asian REITs offer diversification benefits to investors as they have low correlations with other asset classes; (2) the correlation structures are non-constant over time, and (3) there are significant mean and volatility spillover effects from the U.S. REIT market to the REIT markets in Japan, Hong Kong, and Singapore.
The primary purpose of this study is to examine the dynamic transmission of REIT returns and volatility between seven Asian REIT markets: Japan, Singapore, Hong Kong, Malaysia, Taiwan, Thailand, and South Korea from June 2006 to May 2011. The rationale for this research is motivated by a number of reasons. Firstly, since the early works of Markowitz (1952) on portfolio theory, the issue of market integration has become an important subject for both academics and investors. Previous studies on stock markets have provided evidence that markets that are geographically and economically close tend to influence one another and such integration or co- movements is a result of closer political and economic cooperation among countries. Thus, this paper seeks to understand the extent to which a certain movement in one REIT market affects subsequent movements in other REIT markets in Asia. Secondly, most of the previous studies mainly focused on the interactions between the U.S. and major Asian REIT markets, namely Japan, Hong Kong, and Singapore, while little attention has been paid to emerging markets, such as Malaysia, Taiwan, Thailand, and South Korea.
Recent years have witnessed these four emerging markets enjoying rapid growth in market capitalization underpinned by strong economic growth and favorable regulatory changes. Therefore, the linkages of these emerging markets with developed markets in the region deserve closer attention. This study aims to answer the question on whether developed REIT markets influence emerging REIT markets in Asia. Thirdly, since a number of previous studies on stock markets found that international stock market correlations tend to increase during a financial crisis, this study examines whether REIT markets also experience the same phenomenon. In addition, the timeframe will be broken down into three sub-periods in order to analyze and compare REIT market interactions prior to, during, and after the recent Global Financial Crisis (GFC). Therefore, this study makes the following contributions to the existing literature on Asian securitized real estate: (1) provision of empirical evidence on the nature of information transmission and market linkages between Asian REIT markets at both the mean return and volatility levels; (2) extension of the current literature to four emerging REIT markets in Asia: Malaysia, Taiwan, Thailand, and South Korea; and (3) provision of updated evidence on the effects of the GFC on the performance of REIT markets in Asia.
The remainder of the paper is organized as follows. Section 2 provides details of the data and empirical methodology. Section 3 reports the test results on mean and volatility spillovers between the seven Asian REIT markets. Section 4 concludes the paper.
The data employed in this study are the daily closing prices for REIT indices of seven REIT markets in Asia: Japan (JP), Singapore (SG), Hong Kong (HK), Malaysia (ML), Taiwan (TW), Thailand (TL), and South Korea (KO). The indices used are the Standard and Poor’s REIT indices for Japan, Hong Kong, Singapore, and Taiwan, respectively. As REIT indices for Malaysia, Thailand, and South Korea are unavailable, value-weighted indices are constructed by the author to represent listed REITs of each market. In this study, Japan, Hong Kong, and Singapore are classified as developed REIT markets while the rest are categorized as emerging markets. Subsequently, three value-weighted indices are developed to represent developed, emerging, and Asian REIT markets, respectively. The Asian REIT index1 consists of all listed REITs across seven markets as of May 2011. To avoid exchange rate disparities, all indices are commonly measured in term of US$. One of the key characteristic of REITs is that they have to distribute a majority of their net earnings to unit holders. Therefore, dividends are adjusted by using the total return indices for all indices. The returns (denoted by Rt) are continuously compounded and calculated as the difference in natural logarithms of the total return index value (denoted by Pt) for two consecutive trading days, i.e., Rt = ln(Pt) — ln(Pt—1). The period under analysis spans from June 20, 2006 to May 24, 2011. The sample period is chosen on the basis that it represents the longest common time period over which data for the seven REIT markets are available. The total number of observations included in the analysis after adjustment for heterogeneity is 1,286. All historical data are collected from Datastream.
Exhibit 1 illustrates the summary statistics for daily returns of the seven markets, as well as statistics testing for normality and autocorrelation over the period from June 20, 2006 to May 24, 2011. The highest average returns are those of Malaysia (0.053%) and Hong Kong (0.053%), followed by Singapore (0.049%) while the lowest are those of South Korea (0.000%) and Japan (0.012%). Average return of the developed REIT index (0.029%) is greater than that of the emerging REIT index (0.018%). In terms of volatility, as measured by the standard deviation, Japan (1.955%) and Singapore (1.877%) are the most volatile markets while Taiwan (0.739%) and Thailand (0.705%) are the least risky markets during the period under investigation. Interestingly, the developed REIT index (1.450%) is more volatile than the emerging REIT index (0.519%) during the full sample period. It would seem that the emerging markets offer lower returns to investors but lower risk as well. In terms of the risk-adjusted basis, the emerging REIT index is superior to the developed REIT index.
To investigate the dynamic transmission of returns and volatility in these Asian REIT markets, the Exponential Generalized Autoregressive Conditional Heteroskedastic (EGARCH) model is utilized. The EGARCH model, which was developed by Nelson (1991), is an extended form of the generalized ARCH model. ARCH-type models have been commonly used in modeling changes in the volatility of financial time series. A complete survey of the theory and application of ARCH models in finance can be found in Bollerslev, Chou, and Kroner (1992), Bera and Higgins (1993), and Pagan (1996), among others. Importantly, Engle and Ng (1993) and Stevenson (2002) have provided evidence in favor of the EGARCH model over other GARCH-type models. The advantage of EGARCH specification is that it allows for the testing of asymmetry impacts from positive and negative changes in REIT returns. The EGARCH specification employed in this study is written as follows:
where Ri,t is the natural logarithm difference of the total return price indices and i = 1, 2, 3, 4, 5, 6, 7 (1 = Japan, 2 = Singapore, 3 = Hong Kong, 4 = Malaysia, 5 = Taiwan, 6 = Thailand, 7 = South Korea). For market ith, Ωt—1 is the information set at time (t — 1) and hi,t is the conditional variance. Conditional on Ωt—1, ui,t is assumed to be normal distributed with a zero mean and variance hi,t.
Equation (1) describes the conditional mean returns of seven Asian REIT markets, where they can be affected by historical own innovations (own-mean spillovers) when i = j and cross-market returns (cross-mean spillovers) when i G j. The term βi,j measures the direct effect that a change in return to the jth market would have on the ith market. In order to resolve the problem with serial autocorrelation, an MA(1) term is introduced to capture the dependency of current returns (Rt) on immediate past own errors (ut—1).
The DGFC dummy variable is incorporated in the model to evaluate the effects of the GFC on each market. The dummy variable takes the value one in the period from March 17, 2008 to March 6, 2009 and zero otherwise. While it is still unclear about the exact dates of the start and end of the GFC, the March 17, 2008 is used as the beginning of the GFC in this study, since it was the date Bear Stearns was taken over by J.P. Morgan. The March 6, 2009 is chosen as the end for the GFC, because it was the date that the Asian REIT index rebounded strongly from its trough since the start of the GFC.
The correlation analysis is conducted to provide some preliminary insights into the financial contagion of the Asian REIT markets. The correlation matrix of all series from June 21, 2006 to May 24, 2011 is depicted in Exhibit 4. The results show that all correlation coefficients are statistically significant at the 1% level. Correlation coefficients among Asian REIT indices are relatively low, ranging from 0.139 to 0.419. The comparatively high correlations between Singapore and Hong Kong (0.419) might be caused by strong economic and financial linkages between the two countries as both are the major financial centers of the region. The lowest correlations are that between Thailand and South Korea (0.139), followed by Thailand and Japan (0.149). The relatively low correlations between Asian REIT markets indicate potential benefits from regional diversification across Asian REITs. As the average correlation coefficient of the emerging markets (0.25) is lower than that of the developed markets (0.35), investors could certainly achieve greater within-industry diversification benefits by investing in emerging market REITs rather than developed market REITs.
The results of the Ljung-Box (LB) and heteroscedasticity tests for up to 24 orders are reported in Exhibit 6. The LB Q statistics indicate significant autocorrelations in all seven return series. This justifies the inclusion of the lag terms in the mean equation (1). In addition, there is a strong evidence of ARCH effects in the residuals, as suggested from the ARCH test results. This means that there is significant volatility clustering effects in the REIT return indices. Large changes in REIT returns tend to be followed by large changes, while small changes tend to be followed by small changes. This phenomenon is quite similar to observations in previous studies in financial time series. The presence of time-varying volatility in the series further affirms the suitability of an EGARCH model in analyzing the dynamic volatility of Asian REIT markets.
The estimated coefficients and standard errors for the conditional mean return equations are presented in Panel A of Exhibit 7. The results suggest that the lagged returns from Singapore have a significant return spillover effect (measured by the term βij) to all other Asian REIT markets. The mean returns of Japan influence the markets in Malaysia and Taiwan substantially. Significant bi-directional mean spillover coefficients between Hong Kong and Singapore are detected, confirming the strong market linkage between the two countries, as witnessed in the correlation analysis. Among the emerging markets, there are unidirectional mean spillovers from Thailand and South Korea to Malaysia and bidirectional interactions between South Korea and Taiwan. In general, it is more likely for the mean returns to spill from developed REIT markets (e.g., Japan and Singapore) to emerging markets. This implies that investors were looking for price innovations from leading REIT markets such as Singapore and Japan to price REIT securities in other markets in the region. In addition, the returns of Japan, Hong Kong, Taiwan, and Thailand are highly correlated to their past values (βij when i = j) and past disturbance terms (γi), suggesting that historical innovations in those markets have strong influence over expected returns. In other words, Japan, Hong Kong, Taiwan, and Thailand REIT markets are not completely efficient and there are short-run momentums in REIT price.
As explained earlier, a dummy variable DGFC is incorporated to analyze the effects of the GFC on Asian REIT markets. The values of φi and πi measure the impacts of the GFC on mean returns and volatility of each market, respectively. The results show that the impacts of the GFC are not homogeneous across the region. The crisis significantly reduced the returns of all Asian REIT markets, with the exception of Hong Kong and South Korea. When it comes to conditional volatility, most Asian REIT markets react negatively to the crisis, whereas Thailand and Malaysia respond positively. The results imply that Thailand and Malaysia are subjected more to country-specific risks than global risks. This might be attributed to the fact that these two countries have more restrictive currency control regulations than the others. In addition, the GFC affects REIT markets significantly more in the second moment (volatility) than the first moment (mean returns), as the absolute value of dummy coefficients in the variance are greater than those in the mean equations.
The last decade has witnessed a rapid growth of the REIT sector across Asia. However, the body of literature on REIT markets in Asia has been relatively modest as compared to that of the U.S. and Australia. Within the Asian context, this paper investigates the dynamic transmission of REIT returns and volatility between seven countries in Asia: Japan, Singapore, Hong Kong, Malaysia, Taiwan, Thailand, and South Korea employing the EGARCH modification. In addition, the effects of the GFC on REIT returns and volatility are also investigated using dummy variables. The data employed in this study are the daily closing prices for the total return REIT indices from June 20, 2006 to May 24, 2011. The findings have important implications for international investors in managing Asian REIT portfolios, and developing trading and risk management strategies
CS 340 Milestone One Guidelines and Rubric Overview: For this assignment, you will implement the fundamental operations of create, read, update,
Retail Transaction Programming Project Project Requirements: Develop a program to emulate a purchase transaction at a retail store. This
7COM1028 Secure Systems Programming Referral Coursework: Secure
Create a GUI program that:Accepts the following from a user:Item NameItem QuantityItem PriceAllows the user to create a file to store the sales receip
CS 340 Final Project Guidelines and Rubric Overview The final project will encompass developing a web service using a software stack and impleme
