Managing sovereign debts held by the Eurosystem: Operational and legal constraints

Stefano Micossi 01 April 2022



Editors' note: This column is a lead commentary in the VoxEU debate on euro area reform.

Following the proposal by Avgouleas and Micossi (2021) and Micossi (2021), a number of authors (Amato and Saraceno 2022, Baglioni and Bordignon 2022, Cottarelli and Galli 2021, D’Amico et al. 2022) have engaged in the debate on how to manage the sovereign debt portfolio accumulated by the European System of Central Banks (Eurosystem) as a result of their purchase programmes undertaken to raise inflation in the euro area (the PSPP programme, since 2015) and to provide emergency support to the economy in response to the Covid-19 pandemic (the PEPP programme, since 2020). 

The problem arises because the monetary policy justification for the Eurosystem’s sovereign purchases remain valid only as long emergency economic conditions persist, and inflation does not rise back ‘sustainably’ to its 2% target. The key legal issue here is the borderline between monetary and fiscal policy (European Court of Justice Case Weiss and others, C-493/17). The key policy issue is whether the Eurosystem can continue to hold those sovereigns in its portfolio after the end of the special conditions that justified their purchase, or rather dispose of them as monetary policy conditions normalise. In this latter case, it is unclear that the disposal of those assets will not destabilise financial conditions in some national sovereign markets.

These proposals share the common goal of addressing a public policy problem that cannot be pushed under the rug, but they differ in their specific institutional solutions. An assessment of their consistency with present European institutional arrangements may be useful to assess their practical relevance and bring forward the discussion on euro area governance, as I provided in Micossi (2022).

The future of the Eurosystem’s sovereign portfolio 

The sovereigns held by the Eurosystem represent a potential problem, as was indicated above, to the extent that – once the inflation goal is achieved – the fundamental requisite allowing the ECB to hold those sovereigns legally under the TFEU1 may disappear. If so, bonds purchased since 2015 may have to be sold back onto their private markets, thus potentially unsettling monetary conditions and financial stability. 

In their recent “Combined monetary policy decisions and statement” of 16 December 2021, the ECB Governing Council have seemingly outlined a way out of the policy dilemma that has been described by announcing their intention to reinvest the principal from maturing securities both under PEPP (“for at least until the end of 2024”) and APP (their broad asset purchase programme – “for an extended period of time past the date when it starts raising the key ECB interest rates”). 

This approach, unfortunately, papers over the problem rather than solving it, both on operational and on legal grounds. Operationally, the rapid rise of inflation currently under way is already bringing forward the conflict between the ECB’s twin policy goals of draining excess liquidity from the system while at the same time ensuring the orderly rollover of their sovereign portfolio. 

As a follow-up to the ECB combined decision and statement, Baglioni and Bordignon (2022) have argued that, in the current conditions of excess liquidity, the ECB actually has two independent instruments: the interest rate on their deposit facility – which is the floor of the entire interest rate structure – and their asset portfolio to manage excess liquidity. In their view, in the current circumstances the ECB could raise the rate on the deposit facility (currently at -0,5 %) without any need to restrain the monetary base with open market sales. In a similar vein, Cottarelli and Galli (2022) have considered that the ECB could restrict liquidity by imposing an increase in reserve requirements. 

Both schemes imply that the Eurosystem could tighten liquidity and lending conditions without any open market sale of sovereigns, thus overcoming at least for some time the policy dilemma that I have described above. This, however, does not eliminate the issue of the legality, under the TFEU, of leaving the sovereigns in the Eurosystem after the end of their monetary policy justification.  

Monetary policy versus financial stability

Micossi and Avgouleas (2021) have proposed that the Eurosystem sell (a large share of) its sovereigns to the European Stability Mechanism (ESM), which would pay for them by issuing its own euro-liabilities, which would become the European ‘safe’ asset. In a similar vein, Amato and Saraceno (2022) and D’Amico et al. (2022) have proposed the creation of a ‘European Debt Agency’ (EDA) to remove part or all of the outstanding sovereign debts of its member states from capital markets and substitute them with a European safe asset. The safe asset issued by the ESM or other European debt agency would then become the tool of choice for open market operations by the ECB. 

While these proposals are based on the same economic argument, they differ substantially from a legal standpoint. 

What is sometimes insufficiently recognised is that the new Article 136 TFEU and the ensuing decision to establish the ESM have formalised and assigned to the ESM the new and distinct function of preserving financial stability of the member states and the euro area as a whole. With its Pringle judgement (Case C-370/12), the Court of Justice of the European Union (CJEU) has clarified that the objective of preserving financial stability is “clearly distinct” from the objective of maintaining price stability, which is the primary task of monetary policy. Following the establishment of the ESM, the latter institution was given the powers to raise funds to provide financial assistance to its members, to open precautionary credit lines to its members to meet unexpected financial shocks, and to intervene in the primary and secondary market of sovereign securities of its members. There is little doubt, under the Pringle decision, that the task to preserve financial stability belongs to the ESM and not to the ECB. 

It follows from above that:

(a) The issue of freeing the Eurosystem from the encumbrance of national sovereigns requires urgent attention, to the extent that monetary policy requirements may come increasingly into conflict with the task of financial stability.

(b) The ESM offers the proper instrument to bring about full separation between monetary and fiscal policy at the euro area level by transferring sovereign securities held by the Eurosystem to the ESM, as suggested by Micossi (2021), and rolling them over in an orderly way without any interference with monetary policy goals. 

To the extent that over time ECB open market operations were increasingly undertaken with the securities issued by the ESM, the latter could come to cushion, or somehow ‘dilute’, the direct effects of open market operations on individual sovereign markets.    

Abiding by Article 125 TFEU

Article 125 TFEU states the twin prohibition for the Union and for Member States to “be liable for or assume the commitments of central governments, regional, local or other public authorities, other bodies governed by public law, or public undertakings of any Member State, without prejudice to mutual financial guarantees for the joint execution of a specific project”. This is the famous ‘no-bailout’ clause that represents a keystone of the Maastricht Treaty and the ensuing construction of the single currency – and therefore certainly not one aspect of the institutional set up that may be changed to accommodate expedient proposals to facilitate sovereign debt management by the Member States. Thus, any proposal to overhaul sovereign debt management within the EU and, specifically, the euro area must meet the constraint set by Article 125 or risk practical irrelevance. On this score, it appears that Micossi (2021) and Amato and Saraceno (2022) do pass the test, while D’Amico et al. (2022) do not. 

All three proposals would exploit a supposedly vast untapped demand for a European asset guaranteed in the end by the Commission’s ability to enforce sound fiscal policies by its Member States and ensure that sovereign debts are always repaid. The difference between the three schemes boils down to the fact that in the former two the member states remain fully responsible for their debt obligations, with attendant market risks, while in the latter case default and market risks are transferred to a European institution, i.e., they are mutualised.   

Micossi (2021) has proposed that the ESM buys from the Eurosystem a large share of national sovereign debts in their portfolio (not necessarily only their pandemic-related debt) and roll them over perpetually at market conditions. The scheme could entail a temporary support to one Member State in case of default on its liabilities to the ESM; however, the Pringle judgement has made it clear that even in that case the Member State would remain fully liable for its liabilities, including those temporary supports from common institutions or other member states, therefore respecting Article 125 TFEU. 

Amato and Saraceno’s (2022) proposal ultimately entails complete substitution of national debts by common debt issued by a new institution – EDA – which would become the only lender to national governments (the public sector at large) and would charge to each borrower a risk-based interest rate. The principal would not be repaid but would not be cancelled. The combination between insurance against losses and different risk-based interest costs for the borrower (that would include differential charges for the insurance against losses) seems to exclude all mutualisation of debt obligations, and therefore to respect Article 125. 

D'Amico et al. (2022), on the other hand, have seemingly made an explicit choice to ignore institutional constraints and concentrate the attention on the analytical features of their EDA. Under the scheme, the Member States’ sovereign debts incurred due to the pandemic would be purchased by the EDA at market prices, and then cancelled. Accordingly, those debts would also no longer appear in national accounts. The Member States, however, would remain liable to the EDA for an annual contribution calculated so as to repay the agency of its expenses in managing the common debt share of each country and keep the ‘mutualised’ debt constant as a share of national GDP.2 By having the debt agency purchase those debts and mutualise them as EDA debts, D’Amico et al. (2022) seem in blatant violation of Article 125 TFUE. 


The short conclusion of this column is that the euro area needs a new mechanism to free the Eurosystem of the encumbrance of the sovereigns acquired by the ECB in recent years. Such a mechanism – that would entail foregoing repayment on those securities by rolling them over indefinitely at market conditions – cannot be provided by the Eurosystem itself, since this would eventually be inconsistent with its mandate. The ESM could perform that task while respecting all relevant European law, and notably article 125 TFEU (Micossi 2021). 


Amato, M and F Saraceno (2022), “Squaring the circle - How to guarantee fiscal space and debt sustainability with a European Debt Agency”, SEP Working Paper 1/2022, 19 January.

Avgouleas, E and S Micossi (2021), “On selling sovereigns held by the ECB to the ESM: institutional and economic policy implications”, CEPS Policy Insight, 22 March.

Baglioni, A and M Bordignon (2022), “Debito pubblico nelle mani della Bce: uscirne non è un obbligo”,, 1 February. 

Cottarelli, C and G Galli (2022), “Review of the EU economic governance framework: a focus on the revision of the SGP fiscal rules”, Osservatorio Conti Pubblici Italiani.

D'Amico, L, F Giavazzi, V Guerrieri, G Lorenzon and C-H Weymuller (2022), “Revising the European fiscal framework, part 2: Debt management”,, 15 January. 

Leandro, A and J Zettelmeyer (2019), “The Search for a Euro Area Safe Asset”, PIIE Working Paper 18-3 Revised, February 2019.

Micossi, S (2021), “On the selling of sovereigns held by the ESCB to the ESM: A revised proposal”, CEPS Policy Insight No 2021-17, November 202M1.

Micossi S (2022), “Managing sovereign debts held by the ESCB”, CEPS Policy Insights No 2022-02, February 2022.


1 The Treaty on the Functioning of the European Union. 

2 Analytically, this model is close to Leandro and Zettelmeyer’s (2019) euro area leveraged wealth fund. A puzzling technical question arises here regarding the formula for the determination of national contributions, which is derived from the stability conditions for the pandemic debt/GDP ratio in each Member State (the authors want to make sure that the ratio is stabilised) – but may not in most cases entail a positive service payment since the expression (r-g) has been typically negative in historical experience (the important exception being Italy in certain years).



Topics:  EU institutions EU policies

Tags:  Eurosystem, sovereign debt, euro area, European Stability Mechanism

Director General, Assonime; Honorary Professor College of Europe; Member of the Board of CEPS; Chairman of the School of European Political Economy, LUISS.


CEPR Policy Research