Are There Classical Business Cycles?

The aim of this paper is to test formally the classical business cycle hypothesis, using data from industrialized countries for the time period since 1960. The hypothesis is characterized by the view that the cyclical structure in GDP is concentrated in the investment series: xed investment has typically a long cycle, while the cycle in inventory investment is shorter. To check the robustness of our results, we subject the data for 15 OECD countries to a variety of detrending techniques. While the hypothesis is not con rmed uniformly for all countries, there is a considerably high number for which the data display the predicted pattern. None of the countries shows a pattern which can be interpreted as a clear rejection of the classical hypothesis. Addresses of the authors (send correspondence to the rst author): Michael Reiter Ulrich Woitek Department of Economics Department of Economics Universitat Pompeu Fabra University of Glasgow Ramon Trias Fargas 25-27 Adam Smith Building, E-08005 Barcelona Glasgow G12 8RT e-mail: [email protected] [email protected] We are grateful to Fabio Canova, Jim Costain, Claude Hillinger and Jim Malley for very helpful comments and discussions.