Whether discretionary fiscal policies in industrialized countries act counter- or pro-cyclically and whether their reaction is symmetric or asymmetric over the cycle are still largely unsettled questions. This uncertainty remains even when attention is restricted to euro-area countries, where these questions have important implications for the debate on European fiscal rules. We review the recent empirical literature to explain why the results of the various studies differ so greatly. We find that differences are driven partly by the choices made in modelling fiscal behaviour and in the related notions of fiscal policy cyclicality. Results are also affected by data source and vintage (ex post or real-time). The time period chosen is relatively less important. We conclude that the notion of pro-cyclical fiscal policies often upheld in the debate is not justified by the data. Ex post data suggest either a-cyclicality or weak counter-cyclicality. Real-time information gives clearer indications of counter-cyclical behaviour, especially when we progress from a very simple "core" model to a more complex one, including at least the impact of fiscal rules. As for symmetry or asymmetry, the answer varies with sources of data and time periods. With the more complex model the indications of asymmetric behaviour are more robust. Whenever asymmetry is present, it entails shifts in all the parameters of the fiscal rule and not necessarily in the output gap parameter.
Published in 2009 in: Fiscal Studies, v. 30, 1, pp. 39-72
