Michal Franta
The paper analyzes the macroeconomic effects of fiscal policy shocks in the Czech Republic. The low number of observations available for fiscal variables significantly affects the setup of the analysis. Firstly, a small-scale VAR is considered. Secondly, the model is estimated using Bayesian techniques. Finally, all identification approaches that are currently employed by the literature and that are applicable to the Czech Republic are used. The estimation results suggest that the fiscal policy transmission mechanism in the Czech Republic exhibits some standard features (e.g., a rise in GDP and inflation after unexpected government spending, and an increase in government spending after a positive shock to government revenues). However, the uncertainty associated with the results is substantial. Furthermore, it is discussed how the identification strategy itself may represent an additional source of uncertainty of the results.
JEL codes: E62, H30
Keywords: Bayesian vector autoregression, fiscal shocks identification
Issued: December 2012
Download: CNB WP No. 13/2012 (pdf, 817 kB)