We look at this question both theoretically and empirically. Based on data for developed OECD economies, we do find evidence that countercyclicality in fiscal surplus has indeed a significant positive impact on long run productivity growth. However when we decompose the stabilizing pattern of fiscal surplus between government revenues and government expenditures, what comes out of the empirical analysis is very close to Blanchard's point; countercyclical government expenditures have a significant positive impact on growth but countercyclical government receipts have no significant effect on growth. As a matter of fact, following our empirical analysis, it seems that the best way to promote long run growth, and hence to fight the on going economic slowdown is probably to increase government spending, not to reduce government taxes! For those of you who are interested in having a snapshot presentation of the paper and who will be in San Fransisco for the next AEA meetings, the paper will be presented on Sunday, January 4, 2009 at 10:15 AM in the session "Empirics of Appropriate Growth Policy" http://www.aeaweb.org/assa/2009/index.php For those of you who will not be in San Fransisco and/or who would prefer to spend some time reading the paper, I should post the revised version in the coming days. In both cases, please let me know about your reactions! thanks. Enisse