Introduction

Fiscal indicators are the backbone of effective fiscal policy making, including the coordination and surveillance of budgetary policy at the EU level. The quality and success of the EU surveillance framework, in particular the timeliness and appropriateness of any policy recommendation or decision taken in the context of the Stability and Growth Pact (SGP), crucially depend on the quality of its diagnostic instruments. The right conclusions can only be drawn if the underlying analysis is comprehensive and accurate. Ever since its inception, the history of fiscal surveillance in the EU and of the SGP has

inter alia been characterized by a continuous upgrading of the analytical toolkit, so as to be able to respond to new requirements and challenges with the ultimate goal to provide a robust, consistent and accurate assessment of fiscal policy in the Economic and Monetary Union. The 2005 reform of the SGP has confronted us with a number of important new challenges.

Key requirements of the reformed Pact are made contingent on the prevailing economic conditions in the Member States and are expressed in structural terms, net of cyclical factors and one-off and other temporary measures. As a consequence, the reform has broadened and stepped up the scope for gauging the economic and fiscal performance. As regards the preventive arm of the reformed Pact, the adjustment process towards

sustainable medium-term budgetary positions can be modulated depending on prevailing or expected cyclical conditions. In good times the annual structural adjustment should be higher than the 0.5 per cent of GDP benchmark; it can be less than that in economic bad times. The medium-term objectives (MTOs) themselves are defined in structural terms. Finally, as regards the corrective arm of the reformed Pact, the annual adjustment to be achieved by Member States with an excessive deficit is also expressed in structural terms. Besides the increased focus on structural budget balances, there is another and more

complex element in the reformed Pact with important implications for the assessment of fiscal performance, notably long-term sustainability. The EU Council agreement of March 2005, which underpins the reform of the Pact, while confirming the key role of the 3 per cent and 60 per cent of GDP thresholds of the government deficit and debt, puts additional emphasis on the public finance developments over the long term. It specifically identifies the surveillance of debt and sustainability as one of the main areas where improvements can be made to revive the provisions of the SGP. The increased prominence of both the structural budget balance and the long-term sus-

tainability in the EU fiscal surveillance framework raises a number of straightforward yet taxing questions related to the measurement of economic and fiscal performance. Is the current toolkit sufficiently appropriate with respect to the provisions of the reformed SGP? How accurate are the available estimates of the cyclical position and the structural budget

balance? What are the major caveats and how can they be overcome? Besides the available standard indicators, what are the viable alternatives to the assessment of the long-term sustainability of public finances? These and a number of other issues were the focus of the workshop on Fiscal Indicators

in the EU Budgetary Surveillance organized by the Directorate General for Economic and Affairs on 22 September 2006 in Brussels. This volume collects the papers presented at the workshop together with the respective discussions. The three sessions of the workshop mirror the main challenges outlined above, notably (i) long-term sustainability, (ii) the measurement of the underlying budgetary position and discretionary fiscal policy and (iii) the reliability of fiscal indicators. Reflecting the increasing relevance of long-term issues, the first part of this volume covers

two chapters which discuss alternative avenues to assess the long-term sustainability of public finance. The work by F. Ballabriga and C. Martinez-Mongay discusses various tests and rules based on the recent literature on fiscal-reaction functions. They test for a positive response of the primary surplus to accumulated debt in the data and check for robustness considering alternative specifications, estimation techniques and structural breaks. One of their interesting policy conclusions is that stricter conditions do not necessarily ensure sustainability; what matters is enforcement. They conclude that the response to debt has fluctuated over the sample 1977-2005, but sustainability has been prevalent in EU15: most governments have tended to apply a fluctuating but generally positive primary surplus adjustment in response to debt accumulation. The chapter by J. Gokhale uses long-term economic and fiscal projections to determine if

and to what extent current and future expenditure trends, in particular taking into account ageing population, give rise to fiscal imbalances. The results show relatively large gaps which will have to be tackled in the coming years. He proposes the adoption of an extended framework of budget accounting and reporting within the context of the SGP’s long-term fiscal policy surveillance requirement. Gokhale argues that traditional debt and deficit measures can be potentially misleading as measures of a nation’s fiscal stance and suggests adopting fiscal and generational imbalance measures, as these indicators incorporate comprehensive information about future fiscal prospects under current policies. He also provides estimates which show large fiscal imbalances for most EU countries until 2051. The author acknowledges that different underlying assumptions may generate wide variations in estimates of fiscal and generational imbalances and in their ratios to GDP or tax bases. Yet, rather than an argument against adopting such measures, this implies a need to supplement those estimates with measures of the associated uncertainty. The second part of this volume is devoted to the measurement of the underlying budgetary

position and discretionary fiscal policy. The empirical results in Chapters 4 and 5 highlight a tendency of running pro-cyclical policies, especially in economic ‘good’ times. P. Brandner, L. Diebalek and W. Köhler-Töglhofer estimate an unobserved components model while A. Afonso and P. Claeys apply structural VAR analysis. These two chapters identify a recurring pattern according to which estimated changes in the headline deficit net of cyclical factors were not consistent with the policy objective of stabilizing cyclical swings of aggregate output. More in particular, Brandner and co-authors track fiscal policy behaviour over time by decomposing the observed budget balance into four unobserved components: a core balance, an automatic or built-in fiscal stabilizer component, a component reflecting discretionary fiscal policy responses to the business cycle, and, finally, a component reflecting all other transitory shocks to the fiscal position. Their results for Austria highlight that the revenue side seems to be prone to pro-cyclical responses whereas the

expenditure side shows opposite behaviour. Moreover, during economic downturns, the overall impact of fiscal policy seems to be countercyclical, whereas in periods of economic upturn the impact of automatic stabilizers is nearly neutralized. Afonso and Claeys’ focus is on the relation between the cyclical components of total rev-

enues and expenditures and the budget balance in France, Germany, Portugal and Spain. A disaggregate analysis of fiscal policy in a structural VAR that mixes long-and short-term constraints allows them to look into the transmission channels of fiscal policy and to derive a model-based indicator of structural balance. Their main conclusions are that fiscal slippages are mainly due to reversals in tax policies, which are unmatched by expenditure adjustments. As a consequence, deficits rise when economic conditions worsen but cause a ‘ratcheting up’ in the size of government in economic booms. The Pact has not eradicated these pro-cyclical policies. Bad policies in economic ‘good’ times also contribute to aggregate macroeconomic instability. The result of both teams heralded concerns about the conduct of fiscal policy in recent

years. During the expansion preceding the post-2007 global financial and economic crisis, a number of EU Member States did not take fully advantage of the favourable economic conditions to make progress towards their medium-term budgetary objectives (MTOs). Following old habits, they reduced taxes and increased spending in the wake of revenue windfalls just to find out a little later that additional revenues should have been saved for ‘bad’ times. The third part of these proceedings calls attention to the reliability of fiscal indicators; the

two respective chapters recall and illustrate the margins of uncertainty involved in the available set of fiscal indicators. A critical view of the reliability of the underlying fiscal statistics is offered in Chapter 6 by F. Balassone, D. Franco and S. Zotteri on the basis of episodes of large-scale upward revision in government deficits. They discuss the causes of such revisions and raise attention to the potential for opportunistic accounting, or ‘fiscal window dressing’ in their own words. A permanent and detailed cross-checking of available data – starting with, but going deeper than, the simple comparisons of deficit and changes in debt – should reduce the scope for exploiting existing loop-holes in the reporting of budgetary positions. In Chapter 7, R. Barrell, I. Hurst and J. Mitchell provide estimates of the degree of uncer-

tainty attached to the output gap. Uncertainty about the cyclically adjusted budget deficit is further compounded by uncertainty on the link between the gap and the government accounts. They advocate estimates the measurement of fiscal stance should make explicit the bounds around the available estimates on the cyclically adjusted deficit. Yet those in charge of fiscal surveillance cannot waive the responsibility of taking action, even if taking decisions in real time implies taking into account a certain, and in some cases a high, degree of uncertainty.