Which fiscal union?

Guido Tabellini

24 April 2016



Editor’s note: Originally published on 12 February 2016, this is a chapter from the eBook How to fix Europe’s monetary union: Views of leading economists.  

One of the main lessons of the Global Crisis is that to preserve full financial integration and financial stability the Eurozone needs to build elements of a common fiscal policy (see Obstfeld 2013 and Tabellini 2015 and the Five Presidents’ Report, European Commission 2015). In this column I discuss the principles and priorities of how this could be done.

Realising such a deep transformation would require Treaty changes and constitutional reforms in member states, something that does not appear politically feasible in the near term. Yet, at some point the Eurozone will have to grapple with these issues, and the more thoroughly they are discussed, the sooner they will fill the political agenda. Of course, a large literature already exists on this topic. In what follows, I draw in particular on Ubide (2015) and Sapir and Wolf (2015) – also see Corsetti el al (2015) and Paris and Wyplosz (2014).   


Unlike for the US and other federations that achieved integration at an early stage of state development, all Eurozone countries already have large (arguably too large) government spending and taxation. Under any foreseeable scenario, most of these government functions and capacities will have to remain national. The fiscal union should have a few main purposes and priorities, namely to complement the monetary union in two main ways.

  • By providing an arrangement that allows fiscal stabilisation at the level of the Eurozone as a whole.
  • By providing resources to withstand systemic financial crisis (banking crisis and sovereign debt crisis).

The first point was emphasised also in the Five Presidents’ Report. The Eurozone needs a policy tool with which to manage aggregate demand and stabilisation policies during large Eurozone recessions.  European monetary policy should bear the primary responsibility for cyclical stabilisation during normal times. But in exceptional circumstances, monetary policy alone becomes over-burdened and is constrained by the zero lower bound on nominal interest rates. A major lesson of the Global Crisis is that, in such circumstances, monetary policy should be coordinated with fiscal policy to sustain aggregate demand. Given current deflationary trends, it is quite possible that the zero-bound on interest rates will be a recurrent threat during recessions for years to come. If so, the lack of an aggregate fiscal policy tool will be a major handicap for the Eurozone. A primary goal of the fiscal union should be to remove this handicap. Of course, the common fiscal policy tool should aim at the Eurozone average and only in exceptional circumstances, leaving idiosyncratic national shocks to be dealt with by member states.

The second point is instead much more controversial. The Five Presidents’ Report explicitly rules it out, with the argument that the ESM already performs this function.

The ESM arrangement is certainly a very important step forward, but it is doubtful whether its current structure is adequate to prevent the risk of sudden stops. First, its resources (a maximum lending capacity of 500 billion, about 5% of Eurozone GDP) may be insufficient to deal with large systemic crisis - in many European countries bank assets are several multiples of GDP.

Second, the decision to provide stability support to an ESM member is taken by unanimity and requires prior approval by some national Parliaments.1

This makes it highly uncertain and open ended whether and how the ESM resources would actually be available.

For these reasons, the risk of euro exit and of sudden stops remains a significant concern. To be viable in the long run, the monetary union needs an effective system of risk sharing in exceptional circumstances, such as sudden stops and systemic financial crisis (see Obstfeld 2013). Conditioning such an arrangement on the approval of national political majorities vastly reduces its effectiveness, both ex-ante and in the case of need.

In a recent interesting contribution, Gros and Belke (2015) argue that risk sharing through a well-functioning banking union and capital market union may be sufficient to absorb losses from most  financial crises, without the need of a fiscal union, provided that the a common system of deposit re-insurance is in place. A discussion of this issue goes beyond the goal of this article, but I note the following.

  • First, current arrangements limit the resources needed from the newly constituted Single Resolution Fund in the event of a banking crisis, because they impose very demanding bail-in requirements on bank creditors.

This increases the risk of contagion and domino effects, and may not be a viable solution in the event of systemic banking crises.

  • Second, a truly transnational banking union is unlikely to emerge in Europe even in the long run, if the risk of Euro exit and of sudden stops remains significant.

In the presence of this risk, banks will retain a large home bias even if under common (rather than national) supervision.

A comparison with the US, where state debt is negligible and a fiscal union already exists, is misleading. A fiscal union is a pre-requisite for a well- functioning capital market and banking union. Although it may be true that, once credible elements of a fiscal union are in place, the banking system could evolve so that most of the risk would be shared by financial markets and the losses born in the private sector, rather than by tax-payers.

The two points deliberately do not include another controversial issue, namely whether a fiscal union should also perform some of the risk sharing functions towards individuals that are currently performed by national governments. I think this would be a mistake, at least at the current stage. Risk sharing between countries is required in a monetary union, but should mainly be limited to exceptional circumstances, such as sudden stops and systemic financial crisis, or very large shocks. A European system of unemployment insurance (or other welfare programmes directly insuring individual risks) may be politically or symbolically attractive, but would entail great difficulties.

To avoid moral hazard and permanent transfers between countries, a common system of unemployment insurance would require strict harmonization of national labour market institutions, which is difficult and perhaps undesirable. Such a system would also have small economic benefits, for two reasons. First, all members states already have the capacity to directly insure their own citizens. Second, if countries do not lose market access, they should be able to self-insure against small shocks and business cycle fluctuations. What is most important is to insure against large shocks or events that could entail the loss of market access or threats to financial stability. It is in such circumstances that individual member states are powerless, and that systemic externalities are most threatening.

An implication of the foregoing remarks is that the Eurozone does not need to build a large tax capacity of its own. What it needs is the ability to enforce the collection of transfers from member states, not necessarily in very large amounts at once, but for possibly very long periods of time. This would enable the Eurozone to issue its own debt at times of crisis, or for fiscal stabilisation purposes. As in the proposals by Ubide (2015) and Corsetti et al (2015), this debt would be backed by specific tax revenue collected by member states, but earmarked to service Eurozone debt.

According to Corsetti et al (2015), member states would pledge specific sources of future revenue (such as seigniorage, or a fraction of the VAT, or the proceeds from a recurring wealth tax) for say 50 years. The Eurozone could then issue ‘stability bonds’ backed by these future sources of revenue. The proceeds of the stability bonds would then be used to retire national debt from circulation, until all national public debts have reached 60% of GDP in all member states. The main goal of the proposal is to reduce the fragility of the Eurozone by getting rid of the legacy of high public debts.2 Since the current stock of national debt differs between countries, the stability bonds would be backed in greater proportion by pledges from the currently highly indebted countries that need to retire larger amounts of their debt. This is a drawback of the arrangement, which could undermine the rating of the stability bonds.

Ubide (2015) proposes a similar scheme, except that pledges of future revenues would be the same proportion of GDP for all member states, and the stability bonds would primarily be used for fiscal stabilisation and risk sharing in exceptional circumstances, rather than for national debt reduction. He envisages capping the Stability bonds at 25% of Eurozone GDP. Below I discuss more in detail how this arrangement could work. Like in Ubide (2015), debt would be backed by the same GDP percentage of revenue from all member states, and it would be used mainly as an instrument of aggregate demand or crisis management. 

In this perspective, the Eurozone would exploit its key prerogative of being a super-national institution. In pledging and earmarking specific future sources of revenues to the Eurozone in predetermined amounts, member states would accept an irreversible transfer of sovereignty over those revenues. On their own, member states would not have the commitment capacity to credibly earmark future sources of revenue for, say, a sinking fund designed to retire outstanding public debt, or to back a senior debt instrument in times of emergency. An international agreement (with the associated autarky costs of unilaterally breaking the agreement) would provide this commitment capacity. Such an arrangement would give the Eurozone the ability to access financial markets in favourable terms, without necessarily having developed its own tax capacity.


Achieving a common fiscal policy means first of all having a Eurozone policymaker in charge, say a European Fiscal Institute (EFI) to adapt the name from the precursor of the ECB. Coordination of national fiscal policies will not do because enforcement of coordination is inevitably imperfect, and policy needs to be guided by a EZ perspective rather than by national interests.

The EFI could be the logical evolution of the ESM, as it acquires new functions and additional resources, and as it adapts its governance structure to the new greater responsibilities. Sapir and Wolf (2015) have suggested modelling the Fiscal Institute on principles similar to those used for the ECB, clearly a well-functioning European institution.3 The key governing body is the Eurogroup (i.e. a council of national economic and finance ministers that already acts as the Board of Governors of the ESM) plus a smaller executive committee with agenda setting powers and consisting of appointed individuals. The Chairman of the Institute is also the chairman of the executive committee, like the ECB President, and would resemble a Eurozone treasury minister. Most decisions are by simple or qualified majority, depending on the subject matter, with executive committee members having considerable weights.

The key innovation here is the abandonment of unanimity in most decisions. This is inevitable, because if national vetoes can block the implementation of a common policy, then the Fiscal Institute would not be very different from the Eurogroup, and we have already seen the difficulties of this body in reacting to the Global Crisis. This poses the challenge of how to give democratic legitimacy to the common fiscal policy – unlike monetary policy, fiscal policy also concerns redistribution and cannot be guided exclusively by efficiency criteria. The obvious answer is to involve the European Parliament; all major policy decisions of the Fiscal Institute also have to be approved by an ad hoc committee of the European Parliament, or by the European Parliament itself, restricted to Eurozone representatives.

Stability bonds

The main responsibility of the European Fiscal Institute would be to manage a debt instrument of the Eurozone (following Ubide 2015 and Corsetti et al 2015, call it stability bonds), backed by a Eurozone tax capacity. Here is how this could be done.

All EZ member states agree to transfer to the Institute a given amount of their yearly tax revenue (expressed as a percentage of their GDP), upon request by the Institute, up to a pre-established ceiling and up to a pre-determined future date.

The transfer has to be the same percentage for all member states, to avoid redistribution between countries. This would provide the back bones of a Eurozone fiscal capacity with which to service the stability bonds over time. To achieve a high rating on the Stability bonds, member states would also have to give the EFI authority to request at any time an extraordinary transfer of revenues to meet unexpected debt service needs (Ubide 2015).

At the time of issue, the overall amount of Stability Bonds cannot exceed a predetermined percentage of aggregate GDP, say 25% of GDP as in Ubide (2015).

Of course, the Institute’s debt ceiling and the ceiling on the pre-committed national funds would have to be mutually consistent. These ceilings could only be changed under unanimity rule. Ubide (2015) and Corsetti et al (2015) provide different numerical hypothesis and also include a discussion of which sources of government revenue (including seigniorage from the ECB) could be most easily pledged.

To achieve liquidity of the new debt instrument, the EFI would start by gradually issuing a minimum amount of Stability bonds (until it has reached up to say 10% of aggregate GDP). The proceeds from the Stability Bonds would be returned to member states, who would have to retire their own national debt.

The main purpose of the Stability bonds, however, would be to give the Eurozone a new instrument for intertemporal aggregate demand management without relying on fiscal policy coordination.  Thus during deep Eurozone recessions, the EFI would issue additional amounts of Stability bonds and give the proceeds to member states (also in proportion to national GDP), who would be free to use them as they deem appropriate. In particular, the debt proceeds from the Stability bonds could be used to enact a counter-cyclical fiscal policy in the Euro area, if necessary in coordination with monetary policy (including so as to replicate the economic effects of “helicopter money”, as suggested by Turner 2015). To achieve some risk sharing between countries against idiosyncratic cyclical fluctuations, debt proceeds could be distributed according to projected or trend GDPs, while transfers would be collected in proportion to actual GDP.

Similarly, in the event of major systemic financial crisis or sudden stops, the EFI could use the debt proceeds (or part of the yearly transfer from member states) to restore financial stability by lending to member states who have lost market access, under strict conditionality, or to supplement national deposit insurance or to directly recapitalise insolvent financial institutions. In this, the EFI would undertake some of the roles currently attributed to the ESM, which would cease to exist and would merge its procedures and activities in the EFI.

Once the outstanding stock of Stability bonds is sufficiently large to be liquid, the EFI would manage it so as to avoid excessive debt accumulation. Thus, in normal times the transfers from member states would be used to retire the Stability bonds that were issued during previous large recessions (or no new transfers would be collected, if the stock of outstanding debt is deemed appropriate).  

This arrangement would have several benefits. The Eurozone would acquire a fiscal policy tool with which to stabilize aggregate demand or to grant emergency lending. Over time the public debt composition in the Eurozone would also become more efficient. The stability bonds would be relatively safe, because they would be senior to national bonds, would circulate in relatively small quantities, be backed by a pool of revenues from several member states, and be managed by a technical body less easily captured by domestic political uncertainties. They could be used by the ECB for QE and by domestic banks to diversify their portfolio, reducing the risk of the bank-sovereign ‘doom loop’ that was at work during the Crisis. At the same time, national debts would become smaller in size (although by only a small amount). Finally, the stronger enforcement capacity of the EFI compared to current arrangements would give more credibility to the goal of debt reduction in the highly indebted countries, and would more easily prevent new accumulation of national public debts (more on this below).

One drawback of this arrangement is the following. In normal circumstances the stability bonds would be serviced by drawing on transfers from member states up to a pre-determined percentage of GDP. This would probably not be enough to achieve a high credit rating, however. For instance, suppose that a member state defaulted on its obligations because it left the EU. In such extreme circumstances, the predetermined transfers by the remaining member states could be insufficient to service the Stability bonds.  Alternatively, major economic and financial shocks could have similar implications, even if no member states was in default. Hence, to achieve a high credit rating, the EFI should also have the authority to request exceptional transfers in excess of the predetermined amount, if this was motivated by exceptional debt service needs, and according to pre-specified procedures.

This more open ended commitment of resources to back Eurozone debt should be accompanied with a greater ability of the EFI to interfere with national budgetary policy, as described below.

Supervision of national debt policies

Besides managing the common fiscal policy, the European Fiscal Institute would also assume the role currently performed by the European Commission together with the European Council of enforcer of fiscal discipline in member states. With the stronger risk sharing capabilities discussed above, moral hazard would be an even bigger concern than under current arrangements. Moreover, care must be taken to avoid the danger that stability bonds would pile up on unsustainable national debts, rather than leading to an overall debt reduction. Thus, in exchange for the enhanced risk sharing capabilities, member states would have to accept a more intrusive external interference in national fiscal policy.

Specifically, the Fiscal Institute should also have authority to veto national budgets, and impose specific targets for deficits or surpluses, as suggested by Sapir and Wolf (2015). This interference with national political decisions would have to be justified by exceptional circumstances, such as a country being in gross violation of the debt sustainability requirement. The main goal here is to insure adequate fiscal discipline in all member states, but it is also conceivable that the European Fiscal Institute could impose a more lax fiscal policy than approved at the national level, if a fiscal expansion is justified by a major Euro-wide recession.


Over time, the European Fiscal Institute could evolve into a more accomplished fiscal union for the Eurozone. On the one hand, the fiscal union could evolve so as to fund a small set of European public goods such as border patrols, European infrastructures, a European defence system, scientific research.

On the other hand, the revenue collection system could be improved. Rather than relying on transfers from member states, the Institute could be given its own tax bases (or fractions of national tax bases, such as the VAT), on which to levy its own tax rates. This could have symbolic and political benefits, but it would require some centralised tax collection or monitoring capacity, to avoid moral hazard in the enforcement of tax collections. A true EZ tax would also entail the risk of excessive expansion of public spending at the Euro area level, something to be avoided given the size of national government spending. 

Concluding remarks

The arrangement described above entails two obvious politically obstacles, that may be difficult to overcome at least in the near term. First, countries have to give up sovereignty over a fraction of their tax revenues.

It is important to stress, however, that the sacrifice needs not be large in terms of size of yearly revenue. Stability bonds don’t need to be a large fraction of aggregate GDP in order to insure adequate fiscal stabilisation or to provide risk sharing during emergencies. What is essential is the long time horizon, the pledge to transfer national revenue to the Eurozone should extend for a long period of time. A pledge over several decades can provide adequate backing, even if the yearly transfer is relatively small, provided that the arrangement is credible and lasting. In other words, setting up a fiscal union along these lines entails an important element of irreversibility. Without the expectation of irreversibility, the pledge would lack credibility and the arrangement would fail. But the expectation of irreversibility is at the core of the single currency, and it is meant to distinguish it from a fixed exchange rate regime.

The second political obstacle is that the benefits of this arrangement may not be perceived as symmetric. The weaker and highly indebted countries are more likely to lose market access or to be involved in a sudden stop. The benefits of stability bonds are likely to be greater for them than for the stronger member states. On the other hand, the arrangement also entails a greater expected loss of sovereignty for the weaker or highly indebted member states who are more likely to incur in the veto of the Fiscal Institute over their national budgets. In other words, there is an implicit exchange; in order to enjoy the potential benefits of this arrangements, the weaker member states have to accept a temporary loss of sovereignty if they don’t meet the sustainability requirements established ex-ante. This, of course, is a greater loss of sovereignty than that

Acknowledgements: I thank Massimo Bordignon for helpful comments.


G Corsetti, L Feld, P Lane, L Reichlin, H Rey, D Vayanos, B Weder di Mauro (2015), “A New Start for the Eurozone: Dealing with Debt”, Monitoring the Eurozone 1, CEPR.  

Gros, D and A Belke (2015) “Banking Union as a Shock Absorber: Lessons for the Eurozone from the US”, CEPS, Bruxelles

Guiso, L and M Morelli (2014) “Fixing Europe in the Short and Long Run: the European Federal Institute,” VoxEU.org, 3 November.

European Commission (2015) “Completing Europe’s Economic and Monetary Union, Report by J C Juncker, D Tusk, J Dijsselbloem, M Draghi and M Schulz

Obstfeld, M (2013), “Finance at Center Stage: Some Lessons of the Euro Crisis”, Economic Papers 493, European Commission, Director-General for Economic and Financial Affairs

Paris, P and C Wyplosz (2014) “PADRE: Politically Acceptable Debt Restructuring in the Eurozone”, Geneva Report, Special Report 3, ICMB and CEPR.

Sapir, A and G Wolff (2015) “Euro Area Governance: What to Reform and How to Do it”, Bruegel Policy Briefs, February

Tabellini, G (2015) “The Main Lessons to be Drawn from the European Financial Crisis”, in: The Eurozone Crisis: A Consensus View of the Causes and a Few Possible Solutions, by R Baldwin and F Giavazzi, eBook and VoxEU.org, 7 September.

Turner, A (2015) “The Case for Monetary Finance – An Essentially Political Issue”, paper presented at the 16th Jacques Polak Annual Research Conference, IMF

Ubide, A (2015) “Stability Bonds for the Euro Area”, Peterson International Institute for International Economics N. 15-19.


[1] When the European Commission and the ECB both conclude that a failure to urgently grant financial assistance would threaten the economic and financial sustainability of the euro area, unanimity is replaced by an 85% qualified majority. In this case an additional reserve fund needs to be set up as a buffer (Art 4 of the ESM Statute).  

[2] Paris and Wyplosz (2014) formulate an alternative proposal of debt reduction, based on pledging larger amounts of future seigniorage revenues.

[3] Sapir and Wolf (2015) call it the Eurosystem of Fiscal Policy (EFP). In their proposal, the EFP could impose specific targets (for fiscal deficit or surplus) to all member states, with the aim of achieving an appropriate fiscal stance in the Eurozone as a whole. This would be less effective than a fiscal stabilisation achieved through Stability bonds, however, since the fiscal expansion (or contraction) would only occur in some countries and not necessarily in the countries that need it most. Guiso and Morelli (2014) suggest the creation of a European Federal Institute, but they are less specific about institutional details, and they don’t discuss Euro area debt.



Topics:  EU institutions EU policies Monetary policy

Tags:  stability bonds, fiscal union, European Fiscal Institute

Professor of Economics at Bocconi University and CEPR Research Fellow