Pergamon Journal of International Money and Finance, Vol. 16, No. 2. pp. 255-283, 1997 © 1997 Elsevier Science Ltd. All rights reserved Printed in Great Britain PIh S0261-5606(96)00056-3 11261-56{/6/97 $17.00 + O.I)O International business cycles in theory and in practice MORTEN O RAVN* University of Southampton and CEPt~ Department of Economics, University of Southampton, Highfield, Southampton SO171B J, England This paper investigates whether multi-country international business models can account for international comovements. In the OECD there are substantial positive comovements between many output components, between levels of employment, and between total factor productivities. The standard international business cycle model is not consistent with these comovements. Except for consumption levels, the model implies either negative or very low comovements. This is a robust feature of the theoretical models independently of specific parameter values. One inter- esting finding is that a more general specificationof the technology shock processes lead to substantial comovements but at the cost of another deficiency. (JEL E32, E62). © 1997 Elsevier Science Ltd. In this paper it will be investigated whether currently popular multi-country international business cycle models can account for the cross-country comove- ments apparent in data for the OECD economies. A number of recent papers (e.g. Backus and Kehoe, 1992; Backus et al., 1992, 1995; Blackburn and Ravn, 1991; Danthine and Donaldson, 1993; Stockman, 1990) have highlighted the fact that the statistical properties of national business cycles of industrial countries share many aspects, and that there are important cross-country comovements in, especially, output components, employment levels, and in productivity levels. It is, however, still an open question whether theoretical models meant to capture the open economy aspects of business cycles are consistent with the empirical regularities. This is the question addressed in the present paper. The quantitative analysis will be conducted within standard multi-country international business cycle models. Such models have recently gained *Comments from two anonymous referees are gratefully acknowledged. I wish to thank Fabio Canova, Tryphon Kollintzas, Finn Kydland, seminar participants at the University of Aarhus, at the conference 'Quantitative Aggregate Economics', Bergen, August 1993, at the Center for Economic Policy, Athens, and at the CEPR/ESRC workshop on 'Business Cycles' for comments. The usual disclaimer applies. 255