Studies of United States, Swedish, and Australian cities have obtained mixed results regarding the existence of housing market bubbles. The US estimates imply large bubbles on its two coasts, but not inland, and the Australian and Swedish estimates imply insignificant and modest bubbles, respectively. The results of the present work for three New Zealand cities are most similar to those for Swedish cities. Like evidence from the other countries, real income (employment and the real wage rate), real construction costs, and the real after-tax interest rate all matter. While the estimation methodology is similar to an error correction model, it is believed to be a superior method when the data time series is short. In any event, tests for house price bubbles would seem to be necessary in house price research.
Read full abstract