An Analysis of Housing Market Dynamics: Evaluating Potential Bubbles and Their Implications on Affordability
Izabel Brucaj
DOI: https://doi.org/10.24908/iqurcp17994
2024-09-09
Abstract:Purpose Over the past three years, Canada experienced rapidly increasing inflation. Among the many challenges this brought for Canadians, the issue of the affordability of houses and rents is a fundamental one. The rise in Canada’s ratio of the average house price to average rent has fostered further research regarding how this value has changed during the pandemic and its implications for today. This paper extends the research of Allen Head and Huw Lloyd Ellis's exploration of the speculation that Canada is experiencing a housing bubble. A pricing framework is used to assess whether the growth of house prices in Canadian cities since 1987 can be explained by variations in rents, real interest rates and property taxes, known as the “fundamentals.” The magnitude to which house prices have appreciated is contingent upon how participants in the housing market perceive real interest rates. Methodology Firstly, the relevant data needed to conduct our econometric analysis was collected. Data was updated for 27 central metropolitan areas (CMAs) in Canada from 1987 to 2021. Specifically, some of this raw data included rent prices, mortgage rates, interest rates, inflation rates, average and median income, provincial population, and the Consumer Price Index (CPI). In addition, the MLS price was used. Compiled by the Canadian Real Estate Association (CREA), it is an index that collects monthly statistics for properties sold via the Multiple Listing Service (MLS) and accessed by Canadian realtors. Once the raw data was compiled, we calculated the predicted price using a user cost model and compared it to the actual price. This user cost model considers factors such as rents, property taxes and interest rates. From this, we calculated the price-rent ratio. The predicted price-rent ratio, named “User Cost Model price,” as presented in the red lines in the following graphs, demonstrates what our model predicts for overvaluation in the housing market. Comparatively, the actual price-rent ratio, “MLS Average,” presented as the blue line, depicts what we observe in real-time. We then computed the differences between the predicted measure from our model and the actual observed MLS price for each year and city. This differential we calculated is a time series of “overvaluations’’ and “undervaluations.” Using this data for each city, we used this measure of difference in valuation in 2020 and 2021 and compared it to the average difference in valuation over an updated base period. Results and Conclusions Referencing the series of graphs produced, we observe substantial increases in overvaluations in Ontario cities. For example, St. Catharines, London and Windsor depict sizeable overvaluations, in addition to other cities across Canada, such as Toronto, Montreal, Gatineau, and Vancouver. Comparatively, we observed undervaluations in cities in Quebec, such as Québec City, Trois-Rivières, and Saguenay. Cities in the Prairies and Atlantic, such as Calgary, Edmonton, and Saint John, seem to share a similar result of undervaluation. However, each appears to rise in overvaluation again in 2020. While there are many reasons we can attribute to these trends, for now, we may only speculate what this means. We learn from our analysis that overvaluation continues to grow despite increasing rents. These results are particularly interesting for our study because there was a possibility that these rent increases were actually anticipated by housing market participants, which should have accounted for some overvaluation in the past. However, based on our work, this is not the case. If this were true, the overvaluation gap would have narrowed, not widened, as in our case. With this updated dataset, upcoming researchers will be able to investigate deeper into the reasons behind the widening overvaluation gap in many Canadian cities. As statistics become more publicly available, we can continue to enhance this dataset and examine how government policies and housing market participants affect interest rate expectations.