MODELLING ECONOMIC STRUCTURAL CHANGES WITH APPLICATION TO SINGAPORE PRIVATE HOUSING MARKET

Abstract

Variable (DV) approach and the OC method are compared in both theoretical and practical perspectives. This thesis shows that the OC method is able to detect the structural changes in the market accurately, without any prior information about the changing points or the timing of the external shocks. The changing points indicated by the OC method are consistent with the timing of the following policy changes: 1) The Approved Residential Property Schemes (ARPS) in March 1981; 2) CPF allowed to be used to purchase investment properties in May 1986; 3) HDB resale flat valuation and further CPF liberalisation in April 1993. In the analysis of the individual structural shifts in the market, it is also revealed that the prices of public housing have less influence on the prices of private housing, since the CPF liberalisation in 1986. This diminishing effect makes the pricing of public housing an increasingly ineffective tool to control the property price inflation in Singapore. In the application, the OC method is shown to be superior to the DV approach since the former avoids the problem of arbitrary selection of dummies and requires no prior information about the timing of the policy changes. The OC method introduced here provides a powerful tool to analyse structural changes of other economic variables in Singapore, for example, the COE prices in future. The method can provide new insights of the behaviour of other variables, given its robustness to structural changes.