The authors' purpose in this paper is to isolate the respective contributions of budgetary and monetary policy in Canada and the United States to the behaviour of unemployment rates in the two countries. Their method consists of estimating VAR models and using long-term identification restrictions to perform a structural analysis. Budgetary policy shocks are defined as those VAR innovations that cause a change in the trend of the public-spending-to-GDP ratio, while monetary policy shocks are those innovations that modify the trend of inflation. The major finding is that the comparative evolution of unemployment rates in Canada and the United States is linked primarily to the different budgetary policy choices in the two countries. Monetary policy does have some effect on the behaviour of unemployment rates in the two countries, but it does not explain the divergent trends, the primary cause of the U.S.-Canada unemployment gap that has been apparent since the 1970s. The authors estimate, moreover, that the trend in the Canadian unemployment rate has declined significantly over the last few years.