Fiscal rules as institutional tools for public debt management in the European Union Member States

Following the implementation of the provisions of Council Directive 2011/85/EU into national legislation, the literature on the effectiveness of EU Member States’ budgetary frameworks has mainly focused on analysing the construction of numerical fiscal rules (Mohl et al., 2021). The global economic and financial crisis of 2008–2010 catalysed further reforms, including the existing fiscal rules, which proved to be insufficiently effective. Although the term ‘second-generation’ suggests a significant change in the approach to fiscal rules, these changes have merely shifted the focus to their structure, setting new standards that are resilient to economic shocks. Given the lack of a definition of second-generation fiscal rules in the literature (Eyraud et al., 2018), the basis for theoretical considerations in this area is the classical understanding of numerical fiscal rules, supplemented by the guidance provided by Council Directive 2011/85/EU.

The essence of fiscal rules in quantitative terms is best captured by the most frequently cited definition by Kopits and Symansky, namely that they impose a permanent constraint on fiscal policy by setting reference values for selected fiscal aggregates (indicators) such as the budget deficit, public debt, expenditure or revenue, usually in relation to GDP (Kopits & Symansky, 1998). These limits cannot be changed frequently. Fiscal rules are defined as standards in the form of a legal act or political commitment that are binding for at least three years. According to this approach, medium-term budgetary frameworks or expenditure ceilings, which are treated as multi-annual projections that can be revised annually, are not classified as fiscal rules (Budina et al., 2012). The Directive’s guidelines further indicate that country-specific numerical fiscal rules should specify the purpose and scope of their application, how they should be monitored and the consequences of non-compliance. Additionally, the exit clauses in the rules should identify a limited number of specific circumstances and strict procedures under which temporary non-compliance with a rule is permitted. Adapted to the specificities of each Member State and consistent with the budgetary objectives at the Union level, numerical fiscal rules should form the basis of a strengthened EU fiscal surveillance framework (Council Directive 2011/85/EU).

The literature indicates that effective fiscal rules should have three properties: simplicity, flexibility and enforceability (Kopits & Symansky, 1998). In practice, implementing these ideals in the construction of fiscal rules is challenging (Eyraud et al., 2018). Compared to the first generation, second-generation fiscal rules have become more flexible and enforceable at the expense of simplicity (Schaechter et al., 2010; Schick, 2010). This specificity of the rules was intended to enhance the effectiveness of the fiscal response by providing operational guidelines and a roadmap in times of crisis (Bandaogo, 2020). However, the complexity of the rules and additional constraints may hamper their ability to design and implement fiscal policy effectively. Thus, it is a mistake to adopt an exclusively uncritical approach to fiscal rules because of their potentially negative consequences for the economy and public finances (Kumar et al., 2009; Calmfors & Wren-Lewis, 2011). The imposition of overly restrictive rules may lead to limited investment opportunities in the public sector and the need to review public functions by transferring part of the financing to the private sector or by spinning off relevant public sector units performing public functions (Marchewka-Bartkowiak, 2012). Overly restrictive fiscal rules can also lead to the use of ‘creative accounting’ to maintain power and political reputation, which threatens fiscal transparency (Milesi-Ferretti, 2004; Koen et al, 2006; von Hagen & Wolff, 2006; Buti, Nogueira-Martins & Turrini, 2007).

Empirical research confirms the effectiveness of numerical fiscal rules, provided they are properly designed (Potreba, 1996; Alesina & Perotti, 1996; Kumar et al., 2009; Schick, 2010; Bergman, Hutchuson & Jensen, 2016; Eyraud et al., 2018; Bandaogo, 2020; Afonso & Coelho, 2022; European Commission, 2022a). Fiscal rules contribute to improving budget balances (Debrun et al., 2008; Luechinger & Schaltegger, 2013; Maltritz & Wüste, 2015; Bergman et al., 2016; Landon & Smith, 2017; Burret & Feld, 2018; Caselli & Reynaud, 2020) and reducing public debt levels (Debrun et al., 2008; Luechinger & Schaltegger, 2013; Azzimonti, Battaglini & Coate, 2016; Bergman et al., 2016; Landon & Smith, 2017; Asatryan, Castellon & Stratmann, 2018). Numerical fiscal rules also reduce the propensity to run excessive deficits due to impulsive budgetary decisions (Gupta, Liu & Mulas-Granados, 2016; Badinger & Reuter, 2017; Bonfatti & Forni, 2019; Gootjes, de Haan, & Jong-A-Pin, 2021) and contribute to the countercyclical effect of fiscal policy (Blanchard, Dell’Ariccia & Mauro, 2010; Delong & Summers, 2012; Bergman & Hutchison, 2015; Nerlich & Heinrich Reuter, 2015; Combes, Minea & Sow, 2017; Guerguil, Mandon, & Tapsoba, 2017; Gootjes & de Haan, 2022). While limiting budget deficits, the rules simultaneously build and preserve the fiscal space needed for discretionary decisions required by circumstance (Eyraud et al., 2018; IMF, 2018).

The increasing use of fiscal rules has led international organisations to create synthetic indicators to assess the quality of these rules based on their institutional characteristics. An example of such an index covering all types of numerical fiscal rules (budget balance rules, debt rules, expenditure rules and revenue rules) in EU Member States is the Fiscal Rules Strength Index (FRSI) developed by the European Commission’s Directorate-General for Economic and Financial Affairs. Initially, five factors were considered when constructing the FRSI: the legal basis of the fiscal rule, the degree of discretion in setting or reviewing the objectives of the rule, the bodies monitoring compliance with the restrictions imposed, and the transparency of the application of the rule, assessed through the role of the media (European Commission, 2016). However, this methodology was revised in 2015 due to the significant reshaping of EU Member States’ budgetary frameworks in the aftermath of the global economic and financial crisis. Following the guidelines of Council Directive 2011/85/EU on the structure of second-generation numerical fiscal rules, the FRSI considers eight criteria related to the legal basis of fiscal rules, their design and the institutional setting in which they are implemented (European Commission, 2022b). When calculating the FRSI, numerical fiscal rules are scored according to specific scoring weights assigned to each criterion. The resulting scores are then aggregated to calculate a final assessment of the strength of the impact of a given rule, which is comparable across EU member states (Franek & Postuła, 2019). A similar method of calculating an index to assess the quality of numerical fiscal rules is used by the International Monetary Fund (Kumar et al., 2009; Schaechter et al., 2012).

The main criterion for evaluating the condition of public finances and the quality of fiscal policies in EU Member States is the medium-term budgetary objective (MTO), the benchmark of which is the structural balance, as measured by the cyclically-adjusted budget balance (CAB) indicator (Moździerz, 2015; Mourre et al., 2019b). MTOs are generally set within a defined range between 1% of GDP and balance or surplus, in cyclically-adjusted terms, net of one-off and temporary measures (Regulation 1175/2011 of the European Parliament and of the Council). The exception is the euro area countries, for which a structural deficit floor of 0.5% of GDP has been set in line with the Fiscal Compact. However, if the general government debt ratio remains well below 60% of GDP and the risk of losing the long-term sustainability of public finances is low, the lower bound of the MTO may correspond to a structural deficit of at most 1% of GDP (Treaty on Stability, Coordination and Governance in EMU). MTOs ensure the sustainability of public finances or rapid progress towards such sustainability, while leaving room for fiscal policy decisions, with a particular attention to public investment needs (Regulation EU 1175/2011 of the European Parliament and of the Council).

The cyclically-adjusted budget balance is a hypothetical value indicating what the value of the budget balance would be if real output were equal to potential output. The CAB determines the impact of discretionary fiscal policy on the budget by adjusting the actual budget balance for non-structural elements that depend on the impact of cyclical factors. This indicator considers changes in interest payments, which is important for assessing the long-term effectiveness of public debt management, regardless of whether interest payments are directly influenced by the decisions of the current government or whether the level of these expenditures is due to the nature of past fiscal policies. The CAB is also used to assess the size of the fiscal impulse and to examine whether a fiscal policy is sustainable (Larch & Turrini, 2009).

According to the methodology adopted by the European Commission, based on OECD guidelines (Girouard & André, 2005), the CAB is calculated as the difference between the headline budget balance (B, in percentage of GDP) and the estimated cyclical component, which is the product of the semi-elasticity of the headline budget balance (ε) and the output gap (OG = (Y − Yp)/(Yp), i.e. the distance between actual and potential real GDP) (Mourre & Poissonnier, 2019a), which can be expressed by the following formula: CABt=B−εOGt {{CAB}_{t}}=B-\varepsilon {{OG}_{t}} The literature identifies some limitations of the CAB methodology used by the EU due to its aggregated approach. This indicator uses the output gap as a synthetic measure of the business cycle applied to all budget items. Proponents of the disaggregated approach point out that cyclical adjustment based on the output gap may ignore the variation in income and expenditure across phases of the business cycle (Bouthevillain et al., 2001), price effects relevant in periods of high inflation (Morris & Schuknecht, 2007; Escolano, 2010), or the variability of fiscal elasticities over time (Belinga et al., 2014; Köster & Priesmeier, 2017). However, the aggregated method used by the EU performed relatively well during the global economic and financial crisis compared to disaggregated methods (Mourre et al., 2019a).

Despite many reservations, some dating as early as the 1990s (Blanchard, 1990), the CAB remains one of the key indicators for evaluating the nature of fiscal policy, used in the EU fiscal surveillance framework (Larch & Turrini, 2009; European Commission, 2012, Kuusi, 2018, Mourre et al., 2019b).

The budgetary impact of discretionary fiscal policy plays an important role in maintaining the long-term sustainability of public finances. In this context, the medium- to long-term perspective is linked to the assessment of the solvency of the public finances, which is a key criterion for the effectiveness of public debt management (European Central Bank, 2012). Public debt management is the process of setting and implementing a financing strategy to meet the government’s borrowing needs at the lowest possible cost, over the medium to long term, with accepted risk constraints (IMF & World Bank, 2014). The strategy should also pursue other public debt management objectives, such as ensuring the liquidity, efficiency and transparency of the government bond market and the effective management of the government budget liquidity. The main source of raising capital to finance budget deficits and refinance existing liabilities is long-term government bonds issued on both domestic and foreign markets. Nowadays, the government bond market is undoubtedly the most important strategic market for most economies, especially for EU Member States (Postuła, Klepacki & Alińska, 2018).

When developing a public debt management strategy, it becomes crucial to recognise the relationship between budget deficits and long-term interest rates. The literature indicates a key channel for the impact of budget deficits on interest rates, that is national savings (Baldacci & Kumar, 2010). Budget deficits reduce national savings and increase aggregate demand, leading to an excess supply of public debt and higher real interest rates (Elmendorf & Mankiw, 1998). Although short-term real interest rates reflect cyclical conditions and the stance of monetary policy, and influence real medium- and long-term interest rates, the latter are likely to rise more in response to the anticipated worsening in budget deficits and debt (Blanchard, 1984). Large deficits and debt can also raise concerns about the government’s ability to service the debt, making fiscal stabilisation efforts less of a priority (Baldacci & Kumar, 2010; Combes et al., 2017). Thus, it is important for fiscal authorities to strengthen the national budgetary frameworks by implementing numerical fiscal rules (Kumar et al., 2009; Schaechter et al., 2012).

Empirical research confirms the significant impact of numerical fiscal rules on reducing the costs of sovereign borrowing in both domestic and international financial markets (Thornton & Vasilakis, 2018). It indicates that financial markets react positively to the presence of strong rules on fiscal policy outcomes, supporting the thesis that these rules help to address the problems of growing budget deficits and rising debt (Rommerskirchen, 2015). Well-designed numerical fiscal rules also lead to lower interest rates on government bonds (Afonso & Coelho, 2022) and, consequently, to lower yields and spreads on these assets (Iara & Wolff, 2014; Badinger & Reuter, 2017; Gomez-Gonzalez, Valencia & Sánchez, 2022). The literature also highlights that the most important dimension of the effectiveness of numerical fiscal rules in limiting government bond spreads is their entrenchment in law, followed by enforcement mechanisms (Iara & Wolff, 2014).

Given the institutional dimension of numerical fiscal rules, the literature assesses the relationship between fiscal rules and public governance (Bergman et al., 2016). It is measured by the World Bank’s Worldwide Governance Indicators (WGI), which consider a wide range of institutional structures. The six dimensions of the index (voice and accountability, political stability and absence of violence/terrorism, government effectiveness, regulatory quality, rule of law and control of corruption) describe the principle of good governance and rank individual economies in terms of the quality of institutional governance (Kaufmann et al., 2011; Muinelo-Gallo, 2022). This is important in the context of fiscal policy decisions made by government institutions charged with managing public debt, including optimising its servicing costs.

In accordance with the stated aim of this paper, the main research hypothesis has been formulated:

H1: Increasing in the strength of numerical fiscal rules leads to greater effectiveness of public debt management, which is expressed in lower public debt servicing costs.

As numerical fiscal rules interact with the effectiveness of public debt management through different channels, such as the quality of fiscal policy, the solvency of the public finances and the quality of institutional governance, three specific hypotheses have been formulated:

H1a: The higher the value of the CAB, which measures the budgetary impact of discretionary fiscal policy, the lower the public debt servicing costs.

H1b: The lower the solvency of public finances, as measured by the higher yield of 10-year government bonds and the higher the spread of these assets, the higher the public debt servicing costs.

H1c: The higher the quality of institutional governance, expressed in terms of good governance, the lower the public debt servicing costs.

The research was conducted by analysing a sample of 27 EU Member States over the period 2008–2021 using the GMM dynamic panel models estimator. For estimating the effectiveness of public debt management, an analysis of the determinants of the level of debt servicing costs (dependent variable) has been carried out. The explanatory variables included indicators such as the Fiscal Rule Strength Index (FRSI), the cyclically-adjusted budget balance (CAB), the actual budget balance, the general government gross debt, the 10-year government bond yield and the bond spread, as well as the Worldwide Governance Indicators.

To obtain a more detailed understanding of the relationships between variables, subspecifications of the main panel model have been carried out. Within these subspecifications, additional explanatory variables have been included in subsequent iterations to better identify and assess the impact of individual factors on the effectiveness of public debt management. It has been confirmed that the statistically significant relationships obtained are not sensitive to the model specification, i.e. they do not change when additional determinants of public debt servicing costs are included.

The estimation of the models was preceded by an analysis of the correlations between the explanatory variables to identify the interdependencies involved, focusing on the effect of the strength of the numerical fiscal rules on the other explanatory variables. The analysis of the correlation coefficients has revealed a negative correlation between the strength of fiscal rules and the WGI Governance Indicator for Voice and Accountability, Political Stability, and Absence of Violence/Terrorism, and Rule of Law.

The definitions of the explanatory variables used in the estimation of the model are presented in Table 1.

The estimation results of the models in Table 3 indicate that the public debt servicing costs are increasing over time, as evidenced by a positive coefficient with the dependent variable lagged by one period, which is several times higher than the negative coefficient with the dependent variable lagged by two periods. In all model estimations, a negative statistically significant parameter estimate was obtained with the variable reflecting the strength of the numerical fiscal rules. Thus, there are no grounds to reject the main hypothesis H1, which posits a negative relationship between the strength of numerical fiscal rules and the public debt servicing costs. The negative coefficient estimates with the cyclically-adjusted budget balance variable in models (2)–(4) confirm the inverse relationship between the budgetary impact of discretionary fiscal policy and the public debt servicing costs. This confirms hypothesis H1a that the higher the value of the CAB, the lower the public debt servicing cost. Positive statistically significant coefficient estimates with the variables daily 10-year bond yield and bond spread are consistent with hypothesis H1b. The results of the model estimations indicate that the higher the yield of 10-year government bonds and the higher the spread of these assets, the higher the public debt servicing costs.

Higher government gross debt (public debt) and higher actual budget balance (deficit/surplus) are related to higher public debt servicing costs. The estimation results of model (2) indicate that higher political stability and absence of violence are accompanied by lower public debt servicing costs. Only for one of the Worldwide Governance Indicators – Political Stability and Absence of Violence/Terrorism – are there grounds to verify hypothesis H1c. The results of the model estimation are not statistically significant for the other WGIs.