Abstract
This paper investigates the dynamic conditional correlations (DCCs) between housing returns and retail property returns, and the existence of volatility spillover between the two property markets of Hong Kong. Two multivariate stochastic volatility models (MSV), namely Granger causality MSV and DCC-MSV model, are used to capture the time-varying correlations and the volatility spillover effect, respectively. The findings show that the correlations between housing returns and retail property returns follow a dynamic process, and such dynamic correlation could serve as a leading indicator for future property price movements. Besides, the findings also suggest that Hong Kong’s retail property market is generally more volatile than its residential market. Additionally, we find a unilateral volatility spillover from residential property to retail property in the Hong Kong market.
Acknowledgements
The authors wish to express appreciation to the anonymous reviewers for their valuable comments on this paper. This study was funded by the Hong Kong Polytechnic University’s Internal Grants (G-YH86, G-YH96 and Z02Z).
Notes
***1% level of significance.
1. Public sector developments, including domestic units built under the Private Sector Participation Scheme for subsidised sale, and all units built under the Home Ownership, Buy or Rent Option, Mortgage Subsidy, Sandwich Class Housing, Urban Improvement and Flat-for-Sale Schemes are not included.
2. Both the HI and RI are compiled by the Hong Kong government’s RVD, which are the longest most popular property price index in Hong Kong. The methodology of constructing the RVD Index is transaction-based. Specifically, they are based on an analysis made of transactions scrutinised by the RVD for stamp duty purposes. All the data can be obtained via the RVD website http://www.rvd.gov.hk/.
3. The mean value and standard deviations for are 0.2676 and 0.0657, respectively, which suggests that the correlations between the housing and retail price returns are not constant over time.