Abstract/Summary

During housing booms and busts, home prices appreciate and depreciate at very different rates across markets within the same country. However, one thing that is not clear is whether the heterogeneity in house price dynamics we see in markets across a country extends to different municipalities within large metropolitan areas. When a metropolitan area experiences a rapid change in house prices, do house prices deviate from equilibrium homogeneously across the metro area or do some municipalities and house types deviate further than others? Suppose it is the case that home prices deviate heterogeneously across municipalities and house types within a metropolitan area. Does this mean economists are likely to miss the earliest warning signs of overvaluation and a potential housing bubble since any overvaluation in one market segment would be diluted by balanced or undervalued prices in another market segment? Furthermore, what impact does a change in the demand for single-family homes by investors have on house prices across municipalities and house types in the same metropolitan area? The U.S. states that saw the highest volatility in house prices during the U.S. housing bubble of 2006 saw a rapid increase in the demand for homes by investors. Does demand from investors change over time? Do investors gravitate to buying in certain municipalities vs. others, and if so, what impact does this have on house prices? This thesis will address these questions within the context of the largest metropolitan area in Canada. This research uses a dataset of housing sales and rental transactions in the Greater Toronto Area between 2006 and 2019. It isolates 38,561 properties sold and then rented within twelve months of the sale date. This research compares the user costs of owning to the cost of renting each home to test whether a home’s sale price can be explained by fundamental factors or whether the price has deviated from equilibrium. The research found that houses and condos deviated from equilibrium heterogeneously, with condo prices remaining relatively balanced even when houses in the same municipality deviated significantly from equilibrium. Home prices across municipalities showed signs of homogeneity in their direction when deviating from equilibrium but were heterogeneous in the degree of their deviations. Changes in investor demand were more effective at explaining changes in the P/R ratio and deviation from equilibrium for houses than for condos. Finally, this research explored different estimates of user costs to consider cross-sectional differences across house types and price segments. Using a depreciation rate with a declining sliding scale based on the quartile of the home's sale price to account for cross-sectional differences in land value and using different estimates of the maintenance expense for houses vs. condos improved the model’s ability to explain changes in the P/R ratio.