This paper examines the empirical regularity that in Canada business cycle peaks and federal elections have tended to arise together over the long post-Confederation time period following 1867. We argue that rather than being simultaneous, the two events are related sequentially and that causality can be identified properly if the selection issue associated with observed events is addressed carefully. Our results suggest that business cycle peaks lead federal elections rather than the other way around. Such a finding reinforces the hypothesis of strategic election timing for such countries and is insightful in helping to explain why the presence of a political business cycle is harder to establish for parliamentary governments where the date of the next election is under the control of the incumbent governing party than in democratic systems where governing durations and election dates are fixed.

Additional Metadata
Keywords Causality, Election hazard, Election timing, Political business cycles, Selection models
Persistent URL dx.doi.org/10.1016/j.ejpoleco.2012.09.003
Journal European Journal of Political Economy
Citation
Voia, M.-C, & Ferris, J.S. (2013). Do business cycle peaks predict election calls in Canada?. European Journal of Political Economy, 29, 102–118. doi:10.1016/j.ejpoleco.2012.09.003