Social Europe

politics, economy and employment & labour

  • Themes
    • European digital sphere
    • Recovery and resilience
  • Publications
    • Books
    • Dossiers
    • Occasional Papers
    • Research Essays
    • Brexit Paper Series
  • Podcast
  • Videos
  • Newsletter

Reforming Europe’s fiscal rules

Enrico D'Elia 24th March 2020

Why do the Maastricht criteria compare budget deficits and public debt with GDP rather than budget revenues? That’s a good question.

fiscal rules
Enrico D’Elia

The European fiscal rules were conceived about 40 years ago, when the main concern was curbing inflation, and deserve some adjustment, particularly after the Great Recession, since nowadays the main issue is sluggish growth. The Maastricht criteria respectively constrained budget deficits to 3 per cent and public debt to 60 per cent of gross domestic product. The rules should take into account the role of fiscal multipliers and spillovers across the member states, but also measurement issues which plague the debt-to-GDP ratio focused on by the Stability and Growth Pact (SGP).

The first issue has been addressed in the ‘Recommendation for a Council Recommendation on the economic policy of the Euro Area’ released in December 2019, which apparently revives the Keynesian view of the capacity of fiscal stimulus to decrease the debt-to-GDP ratio under many conditions. Particularly, a technical annex reports the multipliers for the main public budget items, estimated for a downturn of GDP and when monetary policy is accommodating.

Fiscal stimulus

These show how even current government expenditure and targeted subsidies have a positive impact on GDP larger than their initial amount, at least in the short run. By contrast, the multipliers of tax are much less than one.

This implies that, when GDP slows, a fiscal stimulus financed by current revenues may accelerate income and reduce the debt-to GDP-ratio. If this is the case, the European fiscal rules must take into account the negative feedback of budget consolidation on GDP and also on the debt-to-GDP ratio, if the latter exceeds approximately the reciprocal of the fiscal multiplier.

Our job is keeping you informed!


Subscribe to our free newsletter and stay up to date with the latest Social Europe content.


We will never send you spam and you can unsubscribe anytime.

Thank you!

Please check your inbox and click on the link in the confirmation email to complete your newsletter subscription.

.

What is more, fulfilling the SGP criteria is counterproductive for debt consolidation if the fiscal multipliers or the debt-to-GDP ratio are large enough. It follows that the limits to debt and deficit cannot be set once and for all, independently of fiscal multipliers—although the commission does not mention this consequence of its recommendation.

Questionable criteria

Another issue, seldom raised by scholars, makes the SGP criteria questionable if the sustainability of debt is the main concern. The public debt may be reimbursed and serviced by withdrawing the required resources from the budget, rather than from GDP. And only public assets could be seized in case of state insolvency, not the private incomes included in GDP. It thus makes more sense to focus on the ratio between debt and public revenues than the conventional ratio to GDP.

The same point holds for evaluating the size of the budget deficit. For instance in 2019, when public revenues were about 46 per cent of GDP in the eurozone, the limits set in the SGP should be restated as a debt not exceeding 129.5 per cent of revenues and the deficit about 6.5 per cent of the same item.

Ratios to revenues do not provide a ranking of member states so different from the traditional measure, as indicated in Graph 1. Only Ireland would slip four positions, possibly because Irish taxation is very generous—particularly for corporations. Yet changing the denominator of the SGP ratios has many intriguing consequences.

Vicious circle

First, public revenues are certified accounting items, while GDP is the results of statistical estimates which are subject to errors and even large retrospective revisions. Thus, a fiscal policy deemed appropriate according to GDP figures published at the time of implementation could turn out to be wrong—if not counterproductive—when GDP data are revised after 3-4 years. During the sovereign-debt crisis this issued in a vicious circle between downward-biased estimates of potential output (derived from misestimates of current GDP) and larger than necessary fiscal adjustments. Such an eventuality would be almost impossible using public revenues as the denominator.


We need your support


Social Europe is an independent publisher and we believe in freely available content. For this model to be sustainable, however, we depend on the solidarity of our readers. Become a Social Europe member for less than 5 Euro per month and help us produce more articles, podcasts and videos. Thank you very much for your support!

Become a Social Europe Member

Graph 1: public debt in the European Union (2019)

fiscal rules
Source: elaboration on AMECO database; (a) excluding UK

Secondly, the measurement of GDP depends on international standards, which can change over time. At the moment, it includes odd items—such as incomes from some irregular activity (firms evading tax, rules and duties) or even criminal actions (prostitution, drug smuggling, trading stolen goods). Eurostat (under)estimated that criminal activities alone account for about 0.5 per cent of GDP in Europe and irregular incomes currently reach about 20 per cent.

It follows that government could be tempted to reduce the debt-to-GDP ratio by allowing the irregular and criminal economy to develop: relaxing laws on drugs, launching a general amnesty or cutting funds for police and justice could all increase GDP. Of course, such policies cannot be recommended and setting deficit and debt targets as ratios of public revenues would make them futile as well. On the other hand, fighting tax evasion, and so raising public revenues, would then help to meet European fiscal targets.

Policy recommendations

Setting the fiscal thresholds in terms of revenues also makes the impact of tax changes on the ratios easy to evaluate. For instance, a tax cut increases the stock of debt and decreases the revenues exactly by the same amount, every time and everywhere, so that it unambiguously raises the debt-to-revenues ratio. The same holds if expenditure cuts compensate the tax cut. Yet the same fiscal measure can have different impacts on debt-to-GDP ratios depending on the fiscal multiplier, which can change across countries, over time and for specific items of the budget. So setting fiscal targets on the revenues ratios also makes fiscal-policy recommendations more clear-cut.

The last two columns of Table 1 report the budget surplus (as percentage of GDP) required in high-debt countries in the next year to reduce by 1/20 the distance between the current debt ratio and that required by the SGP and the revenue criterion respectively. The correction for the business cycle and the other adjustments foreseen by the SGP (such as population ageing) are disregarded for the sake of simplicity.

As expected, the fiscal effort prescribed by the two criteria is not very dissimilar (it is also the same in the eurozone and the European Union as a whole) but it falls slightly below the SGP requirements in five countries out of eight. This markdown may help growth and the political sustainability of debt-reduction plans—in turn a crucial issue for the expectations of financial markets. Notably, a surplus is required by the ratio to revenues in Ireland, because of its exceptionally low revenues-to-GDP ratio, although the Irish debt-to-GDP ratio is below 60 per cent.

Table 1: balance required to reduce the public debt at the pace prescribed by the SGP (2019) (a)

fiscal rules
Source: Elaboration on AMECO database; (a) as percentage of GDP, disregarding the adjustment for the business cycle and other corrections foreseen by the SGP, (b) excluding UK

In sum, setting fiscal targets based on public revenues instead of GDP is not a gimmick to weaken the European fiscal rules, but brings many advantages, in terms of quality, transparency and consistency of policy over time. Hopefully, it could be a step ‘to enhance the effectiveness of the framework in delivering on key objectives’ of the SGP, as sought in the public consultation recently launched by the commission.

Maastricht,fiscal rules,debt-to-GDP ratio,austerity,multiplier
Enrico D'Elia

Enrico D’Elia is an economist who has worked for a number of institutions, including Eurostat as well as the Ministry of Finance, the Statistical Office and the Institute for Economic Analysis of Italy. He has taken part in many co-operation projects and written about a hundred papers on macroeconomic, behaviour of firms and households, inflation and economic forecasting.

Home ・ Economy ・ Reforming Europe’s fiscal rules

Most Popular Posts

schools,Sweden,Swedish,voucher,choice Sweden’s schools: Milton Friedman’s wet dreamLisa Pelling
world order,Russia,China,Europe,United States,US The coming world orderMarc Saxer
south working,remote work ‘South working’: the future of remote workAntonio Aloisi and Luisa Corazza
Russia,Putin,assets,oligarchs Seizing the assets of Russian oligarchsBranko Milanovic
Russians,support,war,Ukraine Why do Russians support the war against Ukraine?Svetlana Erpyleva

Most Recent Posts

Gazprom,Putin,Nordstream,Putin,Schröder How the public loses out when politicians cash inKatharina Pistor
defence,europe,spending Ukraine and Europe’s defence spendingValerio Alfonso Bruno and Adriano Cozzolino
North Atlantic Treaty Organization,NATO,Ukraine The Ukraine war and NATO’s renewed credibilityPaul Rogers
transnational list,European constituency,European elections,European public sphere A European constituency for a European public sphereDomènec Ruiz Devesa
hydrogen,gas,LNG,REPowerEU EU hydrogen targets—a neo-colonial resource grabPascoe Sabido and Chloé Mikolajczak

Other Social Europe Publications

The transatlantic relationship
Women and the coronavirus crisis
RE No. 12: Why No Economic Democracy in Sweden?
US election 2020
Corporate taxation in a globalised era

Eurofound advertisement

Living and working in Europe 2021

The Covid-19 pandemic continued to be a defining force in 2021, and Eurofound continued its work of examining and recording the many and diverse impacts across the EU. Living and working in Europe 2021 provides a snapshot of the changes to employment, work and living conditions in Europe. It also summarises the agency’s findings on issues such as gender equality in employment, wealth inequality and labour shortages. These will have a significant bearing on recovery from the pandemic, resilience in the face of the war in Ukraine and a successful transition to a green and digital future.


AVAILABLE HERE

Foundation for European Progressive Studies Advertisement

EU Care Atlas: a new interactive data map showing how care deficits affect the gender earnings gap in the EU

Browse through the EU Care Atlas, a new interactive data map to help uncover what the statistics are often hiding: how care deficits directly feed into the gender earnings gap.

While attention is often focused on the gender pay gap (13%), the EU Care Atlas brings to light the more worrisome and complex picture of women’s economic inequalities. The pay gap is just one of three main elements that explain the overall earnings gap, which is estimated at 36.7%. The EU Care Atlas illustrates the urgent need to look beyond the pay gap and understand the interplay between the overall earnings gap and care imbalances.


BROWSE THROUGH THE MAP

Hans Böckler Stiftung Advertisement

Towards a new Minimum Wage Policy in Germany and Europe: WSI minimum wage report 2022

The past year has seen a much higher political profile for the issue of minimum wages, not only in Germany, which has seen fresh initiatives to tackle low pay, but also in those many other countries in Europe that have embarked on substantial and sustained increases in statutory minimum wages. One key benchmark in determining what should count as an adequate minimum wage is the threshold of 60 per cent of the median wage, a ratio that has also played a role in the European Commission's proposals for an EU-level policy on minimum wages. This year's WSI Minimum Wage Report highlights the feasibility of achieving minimum wages that meet this criterion, given the political will. And with an increase to 12 euro per hour planned for autumn 2022, Germany might now find itself promoted from laggard to minimum-wage trailblazer.


FREE DOWNLOAD

ETUI advertisement

Bilan social / Social policy in the EU: state of play 2021 and perspectives

The new edition of the Bilan social 2021, co-produced by the European Social Observatory (OSE) and the European Trade Union Institute (ETUI), reveals that while EU social policy-making took a blow in 2020, 2021 was guided by the re-emerging social aspirations of the European Commission and the launch of several important initiatives. Against the background of Covid-19, climate change and the debate on the future of Europe, the French presidency of the Council of the EU and the von der Leyen commission must now be closely scrutinised by EU citizens and social stakeholders.


AVAILABLE HERE

About Social Europe

Our Mission

Article Submission

Membership

Advertisements

Legal Disclosure

Privacy Policy

Copyright

Social Europe ISSN 2628-7641

Social Europe Archives

Search Social Europe

Themes Archive

Politics Archive

Economy Archive

Society Archive

Ecology Archive

Follow us on social media

Follow us on Facebook

Follow us on Twitter

Follow us on LinkedIn

Follow us on YouTube