Diagnosing and Predicting Housing Bubbles: An Application to Provincial Housing Markets in Canada

Akhter Faroque

This paper proposes a methodology for diagnosing and predicting housing bubbles before they burst. The proposal is implemented in two steps. In step 1 we apply three innovative stock-bubbles tests to Canada’s provincial housing markets. This allows us to classify provinces into three categories: those that have not experienced any; those that have experienced in the past; and those that are currently experiencing and may also have experienced in the past episodes of bubble-like explosive growth in the price-rent ratio. To determine whether explosive growth in the latter two categories of provinces is the result of changes in fundamentals or excessive speculation, in step 2 we use the dating chronologies for explosive growth from step 1 and probit models to evaluate the ability of many fundamentals to predict future bubble formation.

DOI : https://doi.org/10.21102/gefj.2017.03.101.03