Abstract
What are the consequences of a fiscal policy measure implemented in a Member State on the rest of the European Union (EU)? Should or should not EU countries coordinate their fiscal policies? Given this starting point, we study the economic consequences of shocks to fiscal variables in the EU countries from both domestic and global perspectives. With that objective in mind, we specify and estimate a global vector autoregressive model (GVAR) for fourteen countries of the former EU15 and the United States (USA), using quarterly macroeconomic, monetary and fiscal data from 1978 to 2009. Unlike other GVAR models with fiscal variables, in our study we consider total public receipts and total public expenditure separately, and model not only the euro area economies but also all countries of the former EU15 (except Luxembourg) and the USA. The results of our simulations show that the responses of real GDP to a negative (positive) domestic/global shock to total public expenditure (total public receipts) seem to be negative (positive) for the analyzed economies. The effects of domestic shocks would be larger in the country of origin of the shock, while their spillover effects would be limited. The effects of global shocks reveal a remarkable degree of similarity in the cyclical behavior of the European economies. As policy recommendations, we suggest boosting the slow process of coordination of fiscal actions in the EU in order to avoid unwanted economic consequences.
Similar content being viewed by others
Notes
In order to assure the validity of the results obtained through the GVAR methodology, four sufficient conditions must hold: (1) acceptance of the weak exogeneity hypothesis of the foreign variables with respect to the long-run parameters of the reduced form for the global vector x; (2) stability of the global model, that is, the eigenvalues of the F matrix defined in equation (3) must be either on or inside the unit circle; (3) smallness of the weights utilized for the construction of the country-specific foreign variables; and (4) weak cross-dependence of the idiosyncratic shocks. For an explanation of the procedure for testing the weak exogeneity hypothesis, see DdPS (2007:12-13), and for a detailed description of the other three conditions for the GVAR estimation, see PSW (2004:137).
Data used in the construction process of time series was mainly taken from the OECD Economic Outlook Database, the OECD Main Economic Indicators Database and the IMF International Financial Statistics Database. In a supplement, available from the corresponding author on request, we provide detailed information on data sources and the construction process of country time series.
Data was taken from the IMF World Economic Outlook Database, September 2011 Edition.
The GVAR Toolbox (Smith and Galesi 2010) used in our calculations generates a lot of intermediate output (e.g. ADF and WS unit root tests, GIRFs with bootstrap error bounds, GFEVDs) in order to check the validity of the results. Because of space constraints, we removed it from the main text. However, the interested reader can obtain a supplement with those and other additional results on request.
The inclusion of the oil price in the US model as an endogenous variable allows the evolution of global macroeconomic variables to have an impact on it. Since the US dollar exchange rate is determined outside the US model, this variable should be treated as weakly exogenous in its specification.
The presence of unit roots in the variables under study invalidates the results obtained from OLS regressions in levels.
See Garratt et al. (2006, Chapt. 6) for a step-by-step explanation on the estimation process of cointegrating VAR models, specifying the different treatments (cases) that can be given to the deterministic components.
The reduced number of available observations in contrast to the large number of parameters to be estimated restricted the specification possibilities of the country-specific models. For this reason, it is satisfactory that only 24 of the 102 regressions do not pass the test of residual serial correlation for the country-specific VECMX* models.
Since the four sufficient conditions for the GVAR estimation are verified, the results of our application are validated. Specifically, (1) the weak exogeneity hypothesis cannot be rejected for 95 out of 101 foreign variables at the 5 % significance level, (2) the model is dynamically stable because the moduli of the 312 eigenvalues of the F matrix in equation (3) are on or inside the unit circle, (3) the trade weights are relatively small, as seen in the trade weight matrix reported in Table 3, and (4) the cross-dependence of idiosyncratic shocks is weak, as emerged from the analysis of the average pair-wise cross-section correlations of the variables of the model and the associated residuals.
See Garratt et al. (2006, Chapts. 6 and 10) for an explanation of the theoretical and practical applications of the GIRFs to VARX* and cointegrating VAR models.
Because of space constraints, simulation results associated with shocks to total public receipts and to total public expenditure are not fully reported here. Those simulation results are provided in a supplement, available from the corresponding author on request.
As pointed by a referee, there are some problems interpreting our results as an assessment of the macroeconomic effects of ‘fiscal shocks’. We recognize such a problem, but given the difficulties in identifying structural shocks in a GVAR model (sign-restriction or narrative identification schemes are not undisputed, and the number of exclusion restrictions and/or the ordering of the variables and the countries in the specification would entail many controversial assumptions; see Caldara and Kamps (2008, 2012), for a comparison of different identification approaches) we decided to summarize the dynamic of the macroeconomic variables in the system following a domestic or area-wide shock through GIRFs. These functions contain not only information about the effects of discretionary fiscal actions, but also information about automatic stabilizers and other effects. Anyway, the GIRFs are informative about the overall macroeconomic effects of the fiscal policy, showing the dynamics of the transmission of shocks in the fiscal variables of one country to itself and also to the other countries in the EU.
See Smith and Galesi (2010) for a detailed presentation of the bootstrap procedure applied to GVAR models.
We should note that the responses of the analyzed variables to the simulated shocks were usually not statistically significant as emerged from the graphs with 90 % confidence bounds (not shown in this paper). This lack of efficiency in the estimates may be due to the relatively small size of the available sample, which forces us to restrict the dynamic specification of the model. In any case, this fact does not detract from an economic interest of our application (as in Galesi and Sgherri 2009, p. 12). On the one hand, it lets us know to what extent the dynamics of the variables among countries are synchronized, following a shock. On the other hand, it enables us to assess to what extent the spillover effects across economies are important, following a shock. In short, we use our GVAR model not so much to quantify but rather to qualify the economic performance of the analyzed countries.
GIRF figures related to this subsection are available in an annex, which can be obtained at https://sites.google.com/site/fiscalgvar/.
GFEVD tables related to this subsection are available in an annex, which can be obtained at https://sites.google.com/site/fiscalgvar/.
References
Beetsma R, Giuliodori M, Klaassen F (2006) Trade spill-overs of fiscal policy in the European Union: a panel analysis. Econ Policy 21(48):639–687
Blanchard OJ, Perotti R (2002) An empirical characterization of the dynamic effects of changes in government spending and taxes on output. Q J Econ 117(4):1329–1368
Caldara D, Kamps C (2008) What are the effects of fiscal policy shocks? A VAR-based comparative analysis. ECB Working Paper Series No 877 / March 2008, European Central Bank. http://www.ecb.europa.eu/pub/pdf/scpwps/ecbwp877.pdf. Accessed 24 June 2009
Caldara D, Kamps C (2012) The analytics of SVARs: a unified framework to measure fiscal multipliers. Finance and Economics Discussion Series 2012–2020, Board of Governors of the Federal Reserve System (U.S.). http://www.federalreserve.gov/pubs/feds/2012/201220/201220pap.pdf. Accessed 8 May 2013
Campa JM (2012) Fiscal challenges in the Euro Zone. Asian Econ Policy Rev 7(2):180–197
De Castro F (2006) The macroeconomic effects of fiscal policy in Spain. Appl Econ 38:913–924
Dées S, di Mauro F, Pesaran MH, Smith LV (2007) Exploring the international linkages of the euro area: a global VAR analysis. J Appl Econ 22(1):1–38
Dickey DA, Fuller WA (1979) Distribution of the estimators for autoregressive time series with a unit root. J Am Stat Assoc 74:427–431
Favero C (2002) How do European monetary and fiscal authorities behave? IGIER Working Paper Series, No. 214, May 2002. ftp://ftp.igier.uni-bocconi.it/wp/2002/214.pdf. Accessed 5 May 2010
Galesi A, Sgherri S (2009) Regional financial spillovers across Europe: a global VAR analysis. IMF Working Paper WP/09/23, International Monetary Fund. http://www.imf.org/external/pubs/ft/wp/2009/wp0923.pdf. Accessed 9 Jan 2010
Garratt A, Lee K, Pesaran MH, Shin Y (2006) Global and national macroeconometric modelling: a long-run structural approach. Oxford University Press, Oxford
Giavazzi F, Pagano M (1990) Can severe fiscal contractions be expansionary? Tales of two small European countries. In: Blanchard OJ, Fisher S (eds) NBER macroeconomics annual. MIT Press, Cambridge, pp 75–122
Hebous S (2011) The effects of discretionary fiscal policy on macroeconomic aggregates: a reappraisal. J Econ Surv 25(4):674–707
Hebous S, Zimmermann T (2010) Budget deficit spillover effects in the Euro Area, (November 3, 2010). Social Science Research Network. http://ssrn.com/abstract=1702291. Accessed 25 July 2011
Hebous S, Zimmermann T (2013) Estimating the effects of coordinated fiscal actions in the Euro area. Eur Econ Rev 58:110–121
Kameda K (2012) Estimating non-Keynesian effects for Japan. Asian Econ Policy Rev 7(2):227–243
Koop G, Pesaran MH, Potter SM (1996) Impulse response analysis in nonlinear multivariate models. J Econ 74:119–147
Leybourne S, Kim T-H, Newbold P (2005) Examination of some more powerful modifications of the Dickey–Fuller test. J Time Ser Anal 26:355–369
Park H, Fuller W (1995) Alternative estimators and unit root tests for autoregressive process. J Time Ser Anal 16:415–429
Perotti R (2005) Estimating the effects of fiscal policy in OECD countries. CEPR Discussion Paper, No. 4842. http://ssrn.com/abstract=717561. Accessed 3 Jan 2009
Pesaran MH, Shin Y (1998) Generalized impulse response analysis in linear multivariate models. Econ Lett 58:17–29
Pesaran MH, Schuermann T, Weiner SM (2004) Modeling regional interdependencies using a global error-correcting macroeconometric model. J Bus Econ Stat 22:129–162
Pesaran MH, Shin Y, Smith R (2000) Structural analysis of vector error correction models with exogenous I(1) variables. J Econ 97:293–343
Ramajo J (2008) Asimetrías y efectos desbordamiento en la transmisión de la política fiscal en la Unión Europea: evidencia a partir de un enfoque VAR estructural. Papel de Trabajo No. 18/08, Institute of Fiscal Studies, Government of Spain. http://www.ief.es/documentos/recursos/publicaciones/papeles_trabajo/2008_18.pdf. Accessed 3 Jan 2009
Sims CA (1980) Macroeconomics and reality. Econometrica 48:1–48
Smith LV, Galesi A (2010) GVAR Toolbox 1.0. University of Cambridge. Centre for Financial Analysis and Policy. http://www-cfap.jbs.cam.ac.uk/research/gvartoolbox/index.html. Accessed 3 Feb 2011
Spilimbergo A, Symansky S, Blanchard O, Cottarelli C (2008) Fiscal policy for the crisis. IMF Staff Position Note SPN/08/01. https://www.imf.org/external/pubs/ft/spn/2008/spn0801.pdf. Accessed 29 Dec 2008
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Ricci-Risquete, A., Ramajo-Hernández, J. Macroeconomic effects of fiscal policy in the European Union: a GVAR model. Empir Econ 48, 1587–1617 (2015). https://doi.org/10.1007/s00181-014-0843-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00181-014-0843-5