The European Commission is in the final stages of drafting a legislative proposal, to be published by the end of April, on reforming the economic governance of the European Union. The proposal will follow on from a plan, published in November (European Commission, 2022), that set out the main lines of the Commission's thinking and gave EU countries a chance to respond.
The stakes for the new proposal are very high. If the Commission ignores the criticisms of EU countries, the reform will not pass before the 2024 European elections. This would leave the fiscal governance of the EU in an extended state of limbo. But if the Commission accommodates EU countries too much, the reform will lose the intellectual strengths of the original plan and may morph into something resembling the present system – a Frankenstein's monster of overlapping rules and procedures.
Most criticism relates to the Commission's plan to replace numerical rules, focused on mandatory minimum reductions of debt or deficits, by a structured process – debt sustainability analysis (DSA) – which takes a broad view of the determinants of debt risks. DSAs look at current debt and deficit levels, but also try to project them into the future, making assumptions about economic growth, current and expected future interest rates, primary balances, implicit liabilities and the maturity structure of the debt stock. DSAs attempt to assess whether debt as a share of GDP is likely to come down or explode, and how fiscal plans and uncertainty about the drivers of the debt influence the chance that debt might be on a rising path. The greater this chance, the stronger the case for fiscal adjustment and/or growth-enhancing reform.
There are three main reasons for using DSAs in the EU fiscal framework, as we have argued previously (Blanchard et al, 2021).
First, using all relevant information to assess debt sustainability makes a lot more sense than relying on a few 'magic numbers' that generally will either fail to pick up debt risks or sound a false alarm.
Second, the main weakness of the current system has been poor implementation. Some of this reflects the unavoidable tension between the EU, represented by the European Commission, and governments that do not internalise the risks of unrestrained deficits for other member states. But some of the reluctance to abide by the rules also reflects their poor design. EU countries are much more likely to undertake fiscal tightening if it is not dictated by a seemingly arbitrary rule, but can be justified based on their specific economic circumstances. This is precisely what the DSA-based approach seeks to do.
Third, though DSAs are far from flawless, we know from experience that using them makes for far better discussions of debt sustainability. The main flaw of DSAs is that their results are only as good as the assumptions they are based on. But these assumptions can be made transparent, and the consequences of different assumptions are easy to illustrate. The International Monetary Fund, for example, uses DSAs to assess debt risks in the context of IMF surveillance, to decide whether countries qualify for loans, and to establish what debt relief is needed when countries cannot pay. All these applications involve discussions or even conflicts of interest: between different departments within the IMF, between the IMF and its members, and between creditors and debtor countries. DSAs make these debates more rational and transparent.
The Commission's approach
The European Commission's November proposal attempts to reconcile a DSA-based approach with a rules-based approach. The treaty-based reference values of 3 percent of GDP for the fiscal deficit and 60 percent for debt, would continue to play a significant role, both in guiding fiscal adjustment and in determining whether to open an Excessive Deficit Procedure (in which the Commission proposes fiscal adjustment measures, potentially backed by sanctions). But the required speed of debt reduction would be based on a Commission DSA. Finally, member states would be required to enact hard, numerical ceilings of non-interest expenditures – net of tax changes and cyclical expenditures – that would be consistent with the debt-reduction objective.
The Commission's proposal is intellectually coherent and greatly superior to the current framework. Nevertheless, we share concerns, including those of the German government, that the DSA-based approach proposed by the Commission might give it excessive discretion (Blanchard et al, 2022). This might come at the expense of even-handedness. For example, under political pressure, the Commission may be less tough in its application of DSA to powerful countries. Even if it applies DSA in the same way to all, powerful countries might be better positioned to push back, exploiting the room for discretion (for example, by arguing over assumptions).
Reflecting concerns of this type, EU finance ministers in March asked that the Commission's proposed debt reduction trajectory "should be based on a common methodology to be agreed that is replicable, predictable and transparent" and that "the Commission technical trajectory should ensure a fiscal effort to put debt on a sufficiently declining path or to maintain it at prudent levels, while preserving the sustainability of public finances and promoting reforms and public investment. To that end, common safeguard provisions to ensure sufficient debt reduction and prevent back-loading of fiscal efforts should be explored." They also called for "further clarifications and discussions" of several questions, including "the appropriateness and design of common quantitative benchmarks to support the reformed framework"1.
The German proposals
Much will depend on how the Commission reacts to concerns set out by Germany in an early-April position paper (a 'non-paper' in EU jargon; to read it, click here), circulated in follow up to the March 2023 finance ministers' statement. Germany's proposals as detailed in the paper would do more harm than good, but the concerns underlying them are largely valid. We think there may be a way to address them that preserves the main strengths of the Commission's plan.
The German government set out six proposals, two of which are described as "common safeguard provisions to ensure sufficient debt reduction and prevent back-loading", and one as a "common quantitative benchmark" that would supplement the proposed framework:
- On top of the expenditure ceilings that follow from the Commission's DSA, Germany proposes an additional rule requiring all countries where debt exceeds 60 percent of GDP to keep the growth rate of expenditure below potential growth. The difference between potential growth and net primary expenditure growth would increase either in the debt-to-GDP ratio, or possibly, in relation to the "public debt challenge" (low/medium/high), identified by DSA.
- In addition, national fiscal plans would be required to "lead to a (sufficient) decline in high debt ratios in each year from the start of the reformed fiscal framework". The speed of decline would again be linked to either the level of debt or the "debt challenge". For example, countries with high debt challenges/debt ratios would be required to reduce their debt by at least 1 percent of GDP each year.
- A "revision clause" would require a revision of the new framework after four years (maximum) "if the reformed framework fails to achieve a reduction in the debt ratios" above 60 percent.
- Public investment under EU programmes (eg funded by NextGenerationEU loans and national co-financing) would not count for the purposes of the proposed expenditure rule. However, the minimum requirements for debt reduction would remain unaffected (ie public investment under EU programmes would still count toward debt creation for the purposes of deciding whether debt declines fast enough).
- The "deficit based" excessive deficit procedure (EDP) would remain unchanged. That is, whenever the deficit criterion is breached, unless any excess over the 3 percent reference value is small and temporary, the Commission would be required to open an EDP, whether or not the member state is sticking to the net expenditure ceilings agreed with the Council. This contrasts with the Commission's November plan, in which compliance with these ceilings would have been considered only a "relevant factor" in deciding whether to open an EDP.
- The "adjustment period" must fall within the current electoral cycle, "to allow the democratic process in Member States to shape their economic policies."
Giving EU-linked investment programmes a special status in the manner proposed in point 4 is also a good idea. It would be a way to mobilise finance for European public goods, as proposed by Lindner and Redeker (2023). And Germany is right to insist that the debt-reduction requirement should remain unaffected. Some investment will automatically meet this criterion, by reducing the debt ratio through its impact on growth and government revenues. Investment that does not may still be good investment. But if a reduction in the debt ratio is necessary, it would need to be offset by lower expenditure, higher taxes or additional growth-enhancing reform.
Point 3 makes sense in principle: if the new framework does not deliver a reduction in debt ratios, it should be revised. But the requirement to revise the framework needs to be decoupled from the mechanical rule, proposed in point 2 of the German non-paper, to immediately reduce all debt ratios above 60 percent (see below for a discussion of that rule). Instead, the trigger for a revision should be a failure of the framework to achieve its aims. Whether this is the case may take more time to observe than just four years.
The purpose of proposal 5 seems to be to create yet another safeguard. If a member state manages to negotiate a weak medium-term fiscal plan that leads to breaches of the 3 percent deficit ceiling ex post (the Commission proposal already rules out ex-ante breaches) than the Commission would be obliged to "prosecute" this case, even if the medium-term fiscal plan is on track. Fair enough.
This safeguard comes at a cost, however. Medium-term fiscal plans that satisfy the 3 percent reference value ex ante in a realistic baseline scenario will only breach it ex post in the event of an economic downturn. The Commission would have to open an EDP, but would conclude that the breach of the deficit value was not the country's fault – for example, it was caused by a natural disaster. Requiring additional fiscal adjustment would not make sense. All the EDP should do in such a case is simply reaffirm the agreed expenditure ceilings. By lumping together fiscally responsible countries that have suffered a shock with fiscal sinners, the reputational penalty associated with the EDP (and hence its effectiveness as an enforcement tool) would be weakened. For this reason, it would be better to submit countries that abide by their expenditure ceilings but nonetheless breach the 3 percent limit to a special surveillance regime, rather than forcing them into the EDP.
We also do not endorse the requirement that the "adjustment period" remain within the electoral cycle (proposal 6). Granted, a specific medium-term fiscal-structural plan consisting of specific expenditure, tax and reform measures, cannot and should not bind future governments. Elections are held precisely to decide on such measures. But the net-expenditure ceilings that constrain these medium-term plans should very much be binding on future governments for the purposes of EU surveillance (and potentially the application of sanctions). Net-expenditure ceilings prevent neither increases in gross expenditures (as long as they are offset by new taxes) nor changes in the composition of expenditure. They are hence consistent with any change in policy that a new government might want to implement, with only one exception: raising the net ceiling. But there is a good reason for that, the same reason for having EU-level rules in the first place: fiscal policy that puts debt sustainability at risk is not just a matter of national concern, but a matter of common concern.
Proposals 1-2, finally, are counterproductive. While they pursue valid aims – reducing excessive discretion and preventing potential abuse – they would undermine a framework that is likely to be more effective than the current one in ensuring debt sustainability, and that would do so at a lower cost to output stabilisation and investment.
Consider first the idea of backstopping the framework's net-expenditure ceilings with an additional rule that requires expenditure to fall in relation to GDP (proposal 1). There is nothing wrong with the idea per se, as long as the alternative rule requires lower (but still meaningful) fiscal adjustment than a properly implemented DSA-based expenditure rule. If this is the case, the additional rule would indeed serve as a backstop, in the sense of curtailing either abuse of the DSA methodology or inappropriate dilution of the required adjustment through bilateral negotiations.
As formulated, however, the German proposal mimics the logic of the framework's DSA-based expenditure rule – to require more adjustment from countries with greater debt challenges – while setting the relationship between debt challenges and the "convergence margin" in an ad-hoc way. This will likely conflict with the result of the DSA-based rule. In particular, the supposed backstop may end up requiring more adjustment than the DSA-based rule even if the latter is properly implemented. If this happens, the framework will lose its efficiency and credibility. National ownership would be undermined. It is hard to imagine that EU governments would pay much attention to a mechanical expenditure rule that is contradicted by the results of a well-implemented DSA.
Consider next the idea to fix a binding lower limit for a necessary decline in the debt ratio of an appreciable magnitude in each year (proposal 2). We assume that this would be prospective rather than ex post (as the latter would be a recipe for the worst kind of procyclical fiscal policy, and inconsistent with the use of an expenditure rule). But even the requirement that debt must fall prospectively "in each year from the start of the reformed fiscal framework" has a near-absurd implication: to reduce the adjustment period envisaged by the Commission proposal – the maximum time that can pass until reductions in the deficit begin to lower the debt ratio – from 4-7 years to zero.
This would be counterproductive, in two respects.
First, it would likely delay the initial application of the framework, which might otherwise go into effect in 2025. According to the April 2023 IMF projections, four EU countries with debt over 60 percent of GDP will still have structural deficits of over 4 percent of GDP in 20232. Three of these would need to undertake fiscal adjustment by 1-3 percent of GDP to get their deficits down to the level that is consistent with falling debt in 2025. This seems unlikely.3 Instead, countries would be given an extended informal adjustment period. During this time, a lame-duck version of the current Stability and Growth Pact would remain in place, with all its problems.
Large future shocks of the type experienced in the last few years would create a similar problem. To deal with shocks, the proposed framework envisages an escape clause that would allow discretionary fiscal stimulus. Cutting the adjustment period to zero after the escape clause is deactivated is a recipe for much more protracted use of escape clauses than under the Commission's proposal. Ironically, this means that under the German proposal, adjustment in the aftermath of large shocks would be subject to less discipline than under the Commission's proposal.
Underlying the counterproductive elements of the German proposal is the view that DSAs are a beast that cannot be tamed and thus must be kept in a cage consisting of multiple overlapping rules and safeguards. This undermines the main strength of the Commission's proposal: to identify debt risks and adjustment needs using an apparatus that, if used properly, is vastly superior to simple numerical rules, and as such is much more likely to be taken seriously by national governments.
But DSAs do not bite, and they certainly can be tamed. Rather than going back to simple numerical rules, the German government – and the Commission – should focus on implementing the Council conclusion that "the Commission trajectory should be based on a common methodology to be agreed that is replicable, predictable and transparent, and should include an analysis of public debt and economic challenges."
In practice, this requires several steps.
- First, all EU countries must be comfortable with the methodology and must know how to operate it. Although there is nothing technically wrong with European Commission's current DSA methodology, this may imply that the common methodology will need to be devised from scratch (eg by the European Fiscal Board). This would remove the worry, pervading the German proposals, that the Commission has an intrinsic advantage and can use it for nefarious purposes.
- Second, at the core of the common methodology must be a tool – a model giving rise to a 'fan chart' of future debt paths – the operation of which does not require any discretion, other than to specify baseline projections for the primary fiscal balance, growth and interest rates. Any abuse would hence take the form of feeding implausible projections into the model.
- Third, the European Fiscal Board and national independent fiscal institutions should be tasked with commenting on the quality and plausibility of the underlying projections.
- Finally, both the tool and the interface used to operate it – which should include the possibility of feeding alternative projections into the model to see how this affects the results – should be made public, allowing the press, think tanks and any interested member of the public to detect improper use.
If additional safeguards are still considered necessary after these steps, the best approach would be to require a minimum reduction in the debt ratio after the end of the adjustment period. The required minimum reduction should be calibrated through simulations involving the agreed methodology, to ensure that it really is a backstop – that is, that it does not call for faster adjustment than justified by the agreed methodology. If the 4-7 year adjustment period is viewed as too generous, a shorter period could be considered. In addition, to prevent backloading, the new legislation should include a "specific revision clause which is anchored in the actual reduction in debt ratios", as suggested in the German paper.
A failure to agree on a new, better set of rules or, even worse, a return to the old rules, would be dangerous for the future, and a major blow to the construction of the EU. It must be avoided at all costs.
The authors are grateful to Grégory Claeys, Rebecca Christie, Zsolt Darvas, Jip Italianer, Johannes Lindner, Francesco Papadía and Lucio Pench for helpful comments on an earlier version of this note.
Blanchard, O., A. Leandro and J. Zettelmeyer (2021) 'Redesigning EU fiscal rules: from rules to standards', Economic Policy 36(106): 195–236, available at https://doi.org/10.1093/epolic/eiab003
Blanchard, O., A. Sapir and J. Zettelmeyer (2022) 'The European Commission's fiscal rules proposal: a bold plan with flaws that can be fixed', Bruegel Blog, 30 November, available at https://www.bruegel.org/blog-post/european-commissions-fiscal-rules-proposal-bold-plan-flaws-can-be-fixed
European Commission (2022) 'Communication on orientations for a reform of the EU economic governance framework', COM(2022) 583 final, available at https://economy-finance.ec.europa.eu/system/files/2022-11/com_2022_583_1_en.pdf
Lindner, J. and N. Redeker (2023) '"It's the politics, stupid" Don't squander this golden opportunity for reforming the fiscal rules', Policy Brief, 27 March, Hertie School Jacques Delors Centre, available at https://www.delorscentre.eu/en/publications/fiscal-rules-reform
1. See Council of the European Union press release of 14 March 2023, 'Economic governance framework: Council agrees its orientations for a reform', https://www.consilium.europa.eu/en/press/press-releases/2023/03/14/economic-governance-framework-council-agrees-its-orientations-for-a-reform/.
2. See IMF World Economic Outlook Database, April 2023, https://www.imf.org/en/Publications/WEO/weo-database/2023/April
3. This paragraph has been modified to correct an arithmetic error in an earlier draft.
This publication does not include a replication package.