To investigate whether a potential buyer will pay a premium on a green residential property (that has been labelled with HK-BEAM or GBC), the transaction records of residential properties with or without HK-BEAM certification and the GBC award were used in the model. In order to nullify the differences in income and geographical characteristics, non-green comparables were selected from the same market segment (same location) of green buildings. This way the model was simplified and also provides a more accurate result on the effect of green features.
Price of the housing property is determined by many independent variables. However, the present study carefully chose only few housing attributes (variables) that have significant impact on housing price and are also not directly influenced by the green features. A semi-log form of HPM was used in the study. The housing price was specified in natural logs and regressed against a set of logged (for those with non-linear relationship) and another set of unlogged variables (for those with linear relationship). The model comprises of nine variables under three broad categories of attributes: structural, environmental and locational.
The proposed HPM model is: Equation 1 [Figure omitted. See Article Image.] where Ln (P) represents the logged residential property price, ß1 ...ß9 are the coefficients to be estimated; ß0 is the constant term and [varepsilon]i the stochastic term. The details of all the variables including definitions and expected signs are summarized in Table III [Figure omitted. See Article Image.].