Get access

The international risk sharing puzzle is at business cycle and lower frequency


  • Corsetti is also affiliated with Rome III and CEPR; Dedola is also affiliated with CEPR. We thank an anonymous referee, our discussant Alessandro Rebucci, Marianne Baxter, Mario Crucini, Mick Devereux, Robert Kollmann, and participants at the conference, ‘International Risk Sharing,’ on 22–23 October 2010 in Brussels, for useful comments; we are grateful to Domenico Giannone for many helpful discussions. Charles Gottlieb provided excellent research assistance. The work on this paper is part of PEGGED (Politics, Economics and Global Governance: The European Dimensions), Contract no. SSH7-CT-2008-217559 within the 7th Framework Programme for Research and Technological Development. Support from the Pierre Werner Chair Programme at the European University Institute is also gratefully acknowledged. The views expressed in this paper do not necessarily reflect those of the ECB, the Bank of Spain, or any of the institutions with which the authors are affiliated. Email: giancarlo.corsetti@


Abstract We decompose the correlation between relative consumption and the real exchange rate in its dynamic components at different frequencies. Using multivariate spectral analysis techniques, we show that, at odds with a high degree of risk sharing, in most OECD countries the dynamic correlation tends to be quite negative, and significantly so, at frequencies lower than two years – the appropriate frequencies for assessing the performance of international business cycle models. Theoretically, we show that the dynamic correlation over different frequencies predicted by standard open economy models is the sum of two terms: a term constant across frequencies, which can be negative when uninsurable risk is large; a term variable across frequencies, which in bond economies is necessarily positive, reflecting the insurance intertemporal trade provides against forecastable contingencies. Numerical analysis suggests that leading mechanisms proposed by the literature to account for the puzzle are consistent with the evidence across the spectrum.


On décompose la corrélation entre consommation relative et taux de change réel en ses composantes dynamiques à différentes fréquences. A l’aide de techniques d’analyse spectrale multivariée, on montre que, en contraste avec un haut degré de partage du risque, dans la plupart des pays de l’OCDE la corrélation dynamique tend àêtre assez négative, et de manière significative à des fréquences de moins de deux ans – les fréquences appropriées pour évaluer la performance des modèles de cycle d’affaires international. Théoriquement, on montre que la corrélation dynamique aux diverses fréquences que les modèles standards d’économies ouvertes prédisent sont la somme de deux termes : un terme constant pour toutes les fréquences, qui peut être négatif quand le risque non-assurable est grand; et un terme qui varie selon les fréquences, qui est nécessairement positif dans les économies équipées de marchés obligataires – reflétant l’assurance que le commerce inter-temporel fournit contre les contingences prévisibles. Une analyse numérique suggère que les mécanismes principaux proposés dans la littérature spécialisée pour expliquer l’énigme sont consistants avec les résultats pour toutes les fréquences.