Panel data modelling and the tax-spend controversy in the euro zone

Abstract

This article attempts to differentiate between the debatable tax and spend, spend and tax, fiscal synchronization and institutional separation hypotheses in order to explore empirically the interplay between public expenditures and public revenues in the Economic and Monetary Union (EMU) member states. For this purpose, panel data models are derived to test the validity of the four hypotheses in EMU countries. A notable characteristic of this article is that the four hypotheses are tested by dividing EMU countries into various subgroups and using disaggregated data for government expenditures and revenues. Seeking for the robustness of the empirical evidence, the panel data methods of Generalized Two-Stage Least Squares (GTSLS) and Generalized Method of Moments (GMM) are accordingly applied to identify the relationship between public outlays and taxation receipts. GTSLS and GMM results strongly support the fiscal synchronization hypothesis implying that budget decision-making is significantly influenced by both government expenditures and revenues components.

Keywords:

• tax-spend controversy
• panel unit roots tests
• panel GTSLS and GMM techniques
• fiscal synchronization

JEL Classification:

• E62
• H11

Acknowledgements

I wish to thank an anonymous referee for useful comments and suggestions. The usual disclaimer applies.

Notes

¹ Payne (2003) presents a detailed survey of the international empirical evidence on the revenue–expenditure nexus. Note that Payne (2003) mentions 53 articles on the subject that investigate the four hypotheses using time series techniques.

² The Peacock and Wiseman view is also known in the literature as the displacement hypothesis because the impact of exogenous disturbances displaces the growth pattern of government expenditure.

³ It should be noted that the spend-tax hypothesis is also in line with the Ricardian equivalence proposition. Along with a different procedure of theoretical analysis, the advocators of the Ricardian equivalence proposition argue that public spending leads to increases in tax revenues by means of the linkage between government borrowing and future tax liabilities. According to the Ricardian equivalence proposition, government borrowing today will lead to increased future tax liabilities. In this way government expenditures undertaken today will contribute to additional tax receipts in the future. For more details on the Ricardian equivalence paradigm, see Vamvoukas (1997a).

⁴ Kitamura (2006) provides a comprehensive description of the empirical framework for different overidentifying restrictions tests.

⁵ For more discussion on this point, see Baltagi (2005).

⁶ For more details on the specific definitions of the variables included in our econometric analysis, see Statistical Annex of European Economy, Autumn (2008).

⁷ See Im et al. (2003), for an analytical discussion on IPS unit root test.

⁸ For a detailed analysis on the computation of W_t-bar statistics, see Im et al. (2003).

⁹ To reserve space IPS test results are not reported here. However, IPS panel unit root tests can be obtained upon request.

¹⁰ PPPs for the 12 EMU countries are annual benchmark results computed by Eurostat. PPPs are the rates of currency conversion that eliminate the differences in price levels between the 12 EMU member states. GDP volume indices based on PPPs converted data indicate only differences in the volume of goods and services produced.