Characterizing the Canadian Financial Cycle with Frequency Filtering Approaches
In this note, I use two multivariate frequency filtering approaches to characterize the Canadian financial cycle by capturing fluctuations in the underlying variables with respect to a long-term trend. The first approach is a dynamically weighted composite, and the second is a stochastic cycle model. Applying the two approaches to Canada yields several findings. First, the Canadian financial cycle is more than twice as long as the business cycle, with an amplitude almost four times greater. Second, the overall Canadian financial cycle is most strongly associated with household credit and house prices. Third, while Canadian house prices are mostly associated with the financial cycle, they are also significantly tied to the business cycle. Lastly, house prices are found to lead the overall financial cycle. These results are generally in line with findings for other countries studied in literature. Additionally, I compare each approach’s proneness to revision and find that both are more reliable, when monitored in real time, than the Basel III total credit-to-GDP gap. Nonetheless, further work is encouraged to investigate more variable combinations and undertake a cross-country analysis since data on systemic financial stress in Canada are limited. It should be noted that since the approaches produce a measure of the financial cycle relative to trend, comparison with level indicators (as those monitored in the Bank of Canada’s Financial System Review) is not straightforward.