The main purpose of this thesis was to analyze the housing market in Sydney, Melbourne and Brisbane over the period between 1983 and 2005. The thesis modeled the determinants of house prices in Sydney, Melbourne and Brisbane in a seemingly unrelated regressions framework. Tests for the house price cycle, property bubbles and net present value (NPV) simulations are conducted across the three capital cities. The thesis employs data for Sydney, Melbourne and Brisbane between 1983 and 2005 both annual and quarterly based on availability.
The empirical results suggest that a restricted seeming unrelated regressions model is the most suitable model to estimate the determinants of house prices (e.g. the number of new residential building, population, income, interest rate and tax on property) across the three cities. House rents are found non-significant determinants of house prices. Following previous literature, Granger-causality tests are conducted to investigate the relationship between the prices in the three cities. Causality is defined in the literature as evidence of house price cycles. House price cycles can worsen house price bubbles. A bubble may be defined as a sharp rise in the price of an asset or a range of assets in a continuous process, with the initial rise generating expectations of further rises and attracting new buyers. The structural shifts and deviation between fundamental and actual values explore the presence of property bubbles in these cities. Evidence of house price bubbles is found for Melbourne. Brisbane has no has property bubbles. In Sydney, the differences between fundamental and actual values show the emergence of a small property bubble. NPV simulations imply all the house markets are sound and profitable with exception of Brisbane when prices are assumed to growth at the median, not mean, rate of capital.