While not as common as budget balance rules, expenditure rules have become widespread among EU member states. Compared to other rules, expenditure rules are considered to promote a better balance between budgetary discipline and macroeconomic stabilisation objectives. They can be more transparent and easier to monitor, as the expenditure aggregate can be more easily understood than for example a structural balance aggregate (Ayuso-i-Casals 2012). For these reasons, an expenditure rule is the one element shared by most of the latest proposals for EU fiscal framework reform (Thygesen et al. 2020, Feld et al. 2018, Darvas et al. 2018, Beetsma et al. 2018), in anticipation of the economic governance review initiated by the European Commission.1
In 2017, 20 expenditure rules were in place in the national legislation of 14 member states, compared to only eight in 2000. These rules cover various levels of government, including regional government and social security (Figure 1). Of those covering general and central governments, half mirror the EU expenditure benchmark, and almost a third are multiannual expenditure ceilings, with varying binding force (Figure 2). In many member states, expenditure rules operate jointly with other national rules, such as budget balance rules and debt rules. To put expenditure rules into perspective, in 2017 there were 115 national fiscal rules in place in the EU member states, of which 62 were budget balance rules, 27 debt rules, and six revenue rules. These rules were in addition to rules operating at the EU level.
Figure 1 Coverage of EU expenditure rules
Figure 2 Design of expenditure rules covering the general and central government
Source: Belu Manescu and Bova (2020).
Based on data on rule design from the European Commission’s Fiscal Governance Database, in this column we address a series of empirical questions surrounding the extent expenditure rules can help address the procyclicality bias of fiscal policy: Are these rules effective in reducing this bias? If so, does the design matter for such impact? Also, is the effectiveness of these rules higher in good times or in bad times? Finally, are these rules more effective in isolation or in conjunction with other rules?
Looking at the universe of expenditure rules in the EU over the 1999-2016 period,2 our results confirm earlier findings (Wierts 2008, Holm-Hadulla 2012) that expenditure rules are effective in reducing the procyclicality of fiscal policy (Belu Manescu and Bova 2020). Not only does the mere presence of expenditure rules reduce the procyclicality bias by half, but better designed rules and/or with wider coverage contribute more to this reduction than weaker rules. The procyclicality bias indicates how much expenditure plans change in response to an unexpected change in economic conditions.3 If fiscal policy is to achieve its macroeconomic stabilisation function, expenditure should not react systematically to short-term revenue shocks, which are often of a cyclical nature. However, in case insufficient budgetary room for manoeuvre has been created in good times, a negative revenue shock may lead to a downward adjustment in expenditure. Recent evidence confirming the countercyclical properties of growth-based spending rules was also provided in the context of EU rules (European Commission 2020).
Expenditure rules help reduce the procyclicality bias of fiscal policy when they are well-designed. A good or strong design depends on a rule having a wide coverage, a strong legal base, independent monitoring or having well-specified consequences for non-compliance. These features are measured by an index developed by the European Commission’s Fiscal Governance Database, which reflects those rule characteristics generally thought to make it more effective. Figure 3 shows how the procyclicality bias decreases as a function of the design strength of the rules, indicating that a stronger design enhances the ability of the rules to mitigate more effectively this bias.4 While expenditure will continue to move in a procyclical way with revenues – i.e. rising with revenue windfalls and falling in downfalls – in the presence of expenditure rules such fluctuations are reduced. In addition, the better designed expenditure rules are, the more they counteract this cyclical fluctuation.
Figure 3 Decreasing procyclicality bias as a function of the design strength of expenditure rules
Source: Belu Manescu and Bova (2020).
Note: The graph shows by how much expenditure increases/decreases to a 1 p.p. unexpected revenue shortfall/windfall measured as the annual change in revenue to GDP ratio, as a function of the strength of the expenditure rule index. Since the fiscal rules index enters the regression with two terms, a standalone variable and an interaction term, the marginal impact of interest, namely the procyclicality bias coefficient, will be a function of the index values. As a result, the procyclicality bias coefficient is illustrated for centiles 60, 65, 70, 75, 80, 85, 90, 95, and 100 of the expenditure rule index distribution. The 95% confidence interval is calculated based on Brambor et al. (2006). The estimates, provided in Belu Manescu and Bova (2020), are robust to endogeneity concerns. The fiscal rules index is calculated as the average across five dimensions (the legal basis, the binding nature of the rule, the nature of the enforcement and monitoring body, the correction mechanism and media visibility), summed over all rules in force weighted by the sector coverage and a penalty in case of a second or third rule covering the same sector. It has a theoretical lower bound of 0 in case there are no rules in force and no theoretical upper bound (in this sample the maximum value of the index is 0.8).
How exactly do expenditure rules contribute to reducing procyclicality? As suggested above, an acyclical or countercyclical fiscal policy is compatible with a situation where expenditure policy does not react systematically to revenue shocks. In this situation, a revenue windfall would trigger some savings, while a downfall would not be accompanied by a reduction in revenues. Thus, by keeping spending in check during good times, expenditure rules help create some buffers to be used in bad times and thus to mitigate the fall in spending triggered by a possible revenue downfall. The main argument underpinning their property of reducing procyclicality relies on the fact that while a large part of revenue is sensitive to economic fluctuations (and would hence react in a procyclical way during shocks), most expenditure components are not. Therefore, expenditure rules can better protect expenditure from the economic cycle, and through this confer either an acyclical or a countercyclical behaviour to the fiscal balance.
Our analysis finds support for this mechanism. In the presence of expenditure rules, expenditure policy shows no systematic reaction to revenue shocks, be it good or bad times, which is, by contrast, the case in countries lacking expenditure rules. This contrasts with earlier studies that found only partial support for the effectiveness of expenditure rules, either in good (Holm-Hadulla et al. 2012) or in bad times (Wierts 2008). The inclusion of the recovery following the 2008-2012 Great Recession as well as the recent widespread adoption of expenditure rules among EU member states therefore appears of critical importance to uncovering the full impact of expenditure rules.
Figure 4 Interaction expenditure rules and budget balance rules
Note: See note to Figure 1 and endnote 3 for a definition of the procyclicality bias. The bars indicate the 95% confidence interval for interacted variables.
Moreover, fiscal policy is least procyclical when expenditure rules operate in combination with budget balance rules (Figure 4). An expenditure rule or a budget balance rule working in isolation is found to have a lower impact on the procyclicality bias than the two of them working together.
To conclude, we provide renewed evidence that expenditure rules are effective in reducing the procyclicality bias of fiscal policy. The more expenditure rules are endowed with essential features and the more they operate in conjunction with budget balance rules, the more effective they are in reducing the procyclicality bias.
Further investigation is warranted to understand better the strengths and weaknesses of national expenditure rules. Going forward, research could focus on the interaction between design and compliance to understand what type of rules (for example, ceilings or growth-based rules) could be more effective. It could also examine the interaction with other type of rules to assess synergies and complementarities Furthermore, as highlighted in Pench et al. (2019), a focus on the interaction with other key elements of fiscal frameworks, such as medium-term plans and independent fiscal institutions would be key to capture what institutional setting would be better for policy stabilisation. Overall, understanding how various expenditure rules have been performing, what properties seem to make them more effective and the way they interact with other rules can serve as a useful input into the ongoing reflection about the architecture of EU and national fiscal frameworks.
Authors’ note: The views expressed in this column represent those of the authors and should in no way be attributed to their employer, the European Commission.
Ayuso-i-Casals, J (2012), “National Expenditure Rules – Why, How and When”, European Economy Economic Papers No 473, European Commission, Brussels.
Beetsma, R, N Thygesen, A Cugnasca, E Orseau, P Eliofotou and S Santacroce (2018), “Reforming the EU fiscal framework: a proposal by the European Fiscal Board”, VoxEU.org, 26 October.
Belu Manescu, C and E Bova (2020), “National expenditure rules in the EU: an analysis of effectiveness and compliance”, Discussion Paper 124, European Commission
Darvas, Z, P Martin and X Ragot (2018), “The economic case for an expenditure rule in Europe”, VoxEU.org, 12 September.
European Commission (2020), “Report on Public Finances in EMU 2019”, Part II on “Performance of spending rules at EU and national level – a quantitative assessment”, Institutional Papers No. 133
Feld, L P, C M Schmidt, I Schnabel, I. and Wieland, V. (2018), “Refocusing the European Fiscal Framework”, VoxEU, 12 September.
Holm-Hadulla, F, S Hauptmeier and P Rother (2012), “The Impact of Numerical Expenditure Rules on Budgetary Discipline over the Cycle”, Applied Economics 44(25): 3287-3296.
Larch, M and S Santacroce (2020), “Tracking compliance with EU fiscal rules: A new database of the Secretariat of the European Fiscal Board”, VoxEU.org, 24 September.
Pench, L, S Ciobanu, M Zogala and C Belu Manescu (2019), “Beyond fiscal rules: How domestic fiscal frameworks can contribute to sound fiscal policy”, VoxEU.org, 14 October.
Reuter, W H (2015), “National numerical fiscal rules: Not complied with, but still effective?”, European Journal of Political Economy 39(C): 67-81.
Thygesen, N, R Beetsma, M Bordignon, X Debrun, M Szczurek, M Larch, M Busse, M Gabrijelcic, E Orseau and S Santacroce (2020), “Reforming the EU fiscal framework: Now is the time”, VoxEU, 26 October.
Wierts, P (2008), “How do expenditure rules affect fiscal behaviour?”, DNB Working Paper 166.
1 Communication from the Commission to the European Parliament, the Council, the European Central Bank, the European Economic and Social Committee and the Committee of the Regions on “Economic governance review” of 5.02.2020.
2 This sample period covers the period for which economic data from the stability and convergence programmes and rule design data from the fiscal governance database were available at the time of the analysis.
3 Specifically, surprises in economic conditions are proxied by the deviation between realized and forecast revenues, measured as annual changes in the total revenue to GDP ratio. Similarly, surprises on the expenditure side are proxied by the deviations between realized and planned expenditures, measured as annual changes in the primary expenditure to GDP ratio. Thus, according to graph 3, the procyclicality bias shows that a 1p.p. unexpected increase in the revenue to GDP ratio will translate into a nearly 0.6p.p. increase in the expenditure to GDP ratio in the absence of any expenditure rules (corresponding to a value zero for the index) and into no increase for very strong expenditure rules (corresponding to high values of the index).
4 European Commission’s Fiscal Governance Database is published here: https://ec.europa.eu/info/publications/fiscal-rules-database_en