Fabio C. Bagliano
Centre for Economic Performance, London School of Economics

Giancarlo Marini
Dipartimento di Economia Politica, Università di Siena

   Recherches Economiques de Louvain/Louvain Economic Review
vol.57, 1991, n.1, p.3-19

Abstract

Neither Granger-causality tests nor innovation analyses can be used to discriminate among competing theoretical models of economic fluctuations. In particular, absence of correlation between monetary innovations and output (a finding usually interpreted in favour of purely "real" business cycle theories) could well be the by-product of successful stabilization policies in models preserving a central role for monetary shocks. The extended Lucas (1973) paradigm proposed in this paper may outperform empirically Real Business Cycle models when changes in policy regimes are taken into account. Although the available empirical results are unable to discriminate among competing thoeries of the cycle, further empirical research, evaluating cross-regime and/or cross-country evidence, may shed further light on these issues.