In March 2025, Germany’s decision to reform its constitutional debt brake triggered a sharp spike in ten-year government bond yields across the euro area as markets priced in the prospect of rising public debt. A month later, 12 EU member states requested the activation of escape clauses under the Stability and Growth Pact (SGP) to accommodate soaring defence spending. These episodes crystallise a tension at the heart of EU fiscal governance: if fiscal rules exist partly to reassure financial markets, what happens when compliance becomes politically inconvenient?

A substantial body of research concludes that well-designed fiscal rules in EU member states are associated with lower sovereign yields or risk premia, acting as a credibility-enhancing mechanism. But remarkably little is known about whether investors price actual compliance with rules. After all, the almost three-decade-long history of EU fiscal rules shows that having rules and following them are two entirely different things. That gap is the motivation for this column.

What we do differently

Most of the existing literature focuses on whether EU member states have fiscal rules, or on how stringent their design is, typically measured by indices like the European Commission’s Fiscal Rules Index (FRI)  (e.g. Heinemann et al. 2014, Afonso and Jalles 2019, Islamaj et al. 2024, Căpraru et al. 2026). We shift the focus to what actually happens: does complying, or failing to comply, with fiscal rules move government borrowing costs? The compliance tracker of the European Fiscal Board (EFB) Secretariat – recently updated to include the year 2025 – offers a useful basis to address the question. First, the tracker is centred on rules that are the same for all euro area countries. Second, the compliance indicator of the EFB Secretariat documents two different states for each country over time: compliance and non-compliance. While this setup does not qualify as a random experiment, it is definitely an improvement compared to studies that do not distinguish between the existence of a rule and compliance. 

As shown in Larch et al. (2023), numerical compliance varies significantly across euro area countries. This cross-country variance can be exploited in a multivariate statistical setting. Evidently, numerical compliance per se may not be completely exogenous and causality is likely to be an issue. However, we share those caveats with most existing papers studying the link between fiscal rules and borrowing costs. To control for temporary or cyclical effects we use average compliance scores. Specifically, we construct three-year moving average compliance indicators for each of the four EU fiscal rules under the Stability and Growth Pact – the debt rule, the deficit rule, the structural balance rule, and the expenditure rule – as well as an aggregate indicator. Each indicator ranges from zero (never compliant in the past three years) to one (always compliant in the past three years). The three-year window is chosen deliberately: it smooths over one-off or cyclical effects and better captures the sustained fiscal behaviour that bond investors may be inclined to price.

The empirical framework is a panel of 20 euro area member states over the period 1999-2025. Three questions guide our analysis: Is compliance associated with reduced bond yields? Is the correlation stronger in highly indebted countries? Is it amplified during financial crises?

What we find

Compliance effects are not uniform across the four EU fiscal rules. The clearest result concerns the deficit rule. Sustained compliance – measured as the three-year moving average mentioned above – is associated with a statistically significant reduction in ten-year sovereign bond yields of around 0.5 percentage points in our baseline specification. This is the only rule for which the first of the three questions listed above (Is compliance associated with reduced bond yields?) can be answered in the affirmative without any shadow of a doubt at conventional levels of statistical confidence. 

The debt rule tells a similar yet somewhat more intricate story.  The compliance indicator per se comes with a positive yet statistically insignificant coefficient. This is not entirely surprising: a country’s debt-to-GDP ratio, which we also include in the analysis, is itself the cumulative result of past non-compliance with the debt rule – the very long memory of compliance, so to speak. In fact, when running our panel regressions without debt rule compliance, the lagged debt-to-GDP ratio remains significant with the expected sign. In other words, a higher (lower) ratio is, on average, associated with a higher (lower) yield on ten-year sovereign bonds. 

Figure 1 Compliance effect on ten-year sovereign bond yields

Note: Panel estimates, 20 euro area member states, 1999-2025

The aggregate compliance indicator, which averages across all four numerical rules of the Stability and Growth Pact, yields no robust and statistically significant effect in either direction, underscoring that aggregation masks heterogeneity.

Why do the cyclically adjusted rules behave differently?

A plausible explanation lies in the nature of the rules themselves. The deficit and debt rules are expressed in nominal terms and are directly observable in real time with virtually no uncertainty. Investors can straightforwardly track whether a country’s deficit is or is expected to be above or below 3% of GDP and whether the debt-to-GDP ratio is increasing or declining. By contrast, the expenditure and structural balance rules are defined in cyclically adjusted terms and depend on estimates of the output gap – unobservable variables that are frequently revised and are inherently uncertain. If bond investors cannot readily verify compliance with such a rule, the credibility signal is potentially weaker or noisier, limiting any yield-compressing effect. Moreover, compliance with the expenditure or structural budget balance rule may go along with a deterioration in the headline deficit and the debt ratio. While in the short run such differences are intended to strip out the effect of the cycle, persisting differences may point to underlying measurement issues, for instance a bias in forecasting key macroeconomic variables.

Debt levels matter – but less than expected

For both the debt and deficit rules, interacting the compliance indicator with the debt-to-GDP ratio suggests a threshold in the 75-80% of GDP range, above which compliance is associated with lower yields. This means that once debt exceeds a certain threshold, rule compliance produces an extra benefit in terms of yields. However, none of these interaction effects achieve strong statistical significance, so we cannot offer a clear answer to the second of the three questions listed above (Is the correlation stronger in highly indebted countries?). 

Financial crises sharpen the deficit and debt signals

The results pertaining to the third question (Is the effect of compliance amplified during financial crises?) are the most striking, although not unexpected. During periods of systemic banking or sovereign debt crisis, compliance with the deficit rule is associated with a reduction in sovereign bond yields of approximately 1.5 percentage points relative to non-compliant peers, a result significant at conventional levels of statistical confidence. Compliance with the debt rule during a crisis is associated with a reduction of around one percentage point, also statistically significant. The interpretation aligns with a signalling mechanism: when markets reassess sovereign risks on the back of economic and financial stress, maintaining compliance with visible, nominal fiscal constraints sends a credible message that markets seem to reward with lower yields.

What this means

Taken together, these findings suggest that financial markets may be selective in how they price fiscal rule compliance. They appear to respond to rules that are simple, transparent, and expressed in readily observable terms. They seem, on average, to pay less attention to compliance with rules whose measurement may be conceptually solid but less straightforward in real time or include elements of judgement and/or uncertainty.

These findings may have practical implications for the EU’s fiscal governance framework. The 2024 reform moved towards a single operational indicator: the growth rate of net national expenditure. Taken at face value, our results indicate that the market credibility of an expenditure-based framework cannot be taken for granted. If investors cannot readily interpret compliance with the new rule, or headline deficit or debt numbers do not improve as projected, in spite of rule compliance (e.g. because economic growth projections underpinning the recommended speed limit on government expenditure turn out weaker than expected), the market discipline channel may remain muted.

A broader implication concerns what compliance actually reflects. As Rommerskirchen (2015) and Heinemann et al. (2014) argue, the choice to comply with fiscal rules is not random; it mirrors the underlying fiscal preferences of governments and voters. Countries with a strong ‘stability culture’ tend both to comply and to enjoy lower borrowing costs, making it difficult to attribute lower yields to compliance per se rather than to the latent preferences that drive both. Hence, while we do control for numerical compliance, endogeneity is the central methodological challenge this study cannot fully resolve.

Author’s note: The views expressed in this column do not necessarily reflect those of the European Commission or the European Fiscal Board. The column draws on Alicia’s recently completed master’s thesis at the College of Europe, Bruges. 

References

Afonso, A and J T Jalles (2019), “Fiscal Rules and Government Financing Costs”, Fiscal Studies 40(1): 71-90.

Căpraru, B, G Georgescu and N Sprincean (2026), “Fiscal Rules, Independent Fiscal Institutions and Sovereign Risk: Evidence From the European Union”, International Journal of Finance & Economics 31(1): 29-45.

Heinemann, F, S Osterloh and A Kalb (2014), “Sovereign Risk Premia: The Link between Fiscal Rules and Stability Culture”, Journal of International Money and Finance 41: 110-127.

Islamaj, E, A S Penaloza and S Sommers (2024), “The Sovereign Spread Compressing Effect of Fiscal Rules during Global Crises”, World Bank Policy Research Working Paper No. 10741.

Larch, M, J Malzubris and S Santacroce (2023), “Numeric Compliance with EU Fiscal Rules: Facts and Figures from a New Database”, Intereconomics 58(1): 32-42.

Rommerskirchen, C (2015), “Fiscal Rules, Fiscal Outcomes and Financial Market Behaviour”, European Journal of Political Research 54(4): 836-847.