International Business Cycles: Evidence from Capital Coefficient Based Measures of Capacity Utilisation

The paper refers to capacity utilisation, applying a short-cut that is sometimes used in business cycle research to yearly GDP and investment data from 1960 to the present for 22 countries. The basic idea is that the empirical short-run fluctuations of the capital output ratio v are mainly due to cyclical changes in capital utilisation. Accepting this, the individual HP filtered long-run trend estimate vt for country i can be used to identify the actual deviation of any respective vt from its ‘equilibrium’ level, which in turn allows to quantify capital utilisation. This method is easy to implement and gives an internationally perfectly comparable measure. Next, principal component analysis is used to extract the common variance of these series for our 22 countries in the sample, resulting in 5 orthogonal factors with eigenvalue > 1. These can readily be interpreted as distinct international business cycle country groups. We can identify the following country clusters: a European group comprising France, Spain and Sweden; a group of German speaking countries plus Italy and Japan; a group of English speaking countries (incl. AUS, but not NZL) plus the Netherlands; a Nordic group KV (Norway and Denmark); a complementary Western European group (Portugal, Ireland); New Zeeland as an isolated Economy. The identification of these clusters is suggested as a starting point for further explorations into the regularities as well as the causes of international cyclical co-movement (or the lack of