A safe asset for Eurozone QE: A proposal

Luis Garicano, Lucrezia Reichlin

14 November 2014



As Europe moves closer to deflation, the ECB is gradually inching towards outright quantitative easing (QE) – increasing the monetary base through purchases of government bonds (Draghi 2014). But undertaking such purchases confronts a problem. There is no Eurozone ‘government bond’ to purchase. Were the ECB to purchase the debt of all member countries, it would end up with a large amount of debt on its balance sheet, making it impossible for a country to default without triggering very large redistribution. Thus, QE would effectively mean the ECB is providing insurance to the private sector from sovereign risk. This would eliminate market discipline, as the threat of debt restructuring would be eliminated.

A proposal

Potential alternatives to this would be using only top-rated countries for QE (which would drastically limit the impact of QE) or using some kind of conditional QE like previous ECB programs (unfeasible for the purposes of QE).

We propose here a different alternative – use of a synthetic safe bond formed by the senior tranches of a set of national bonds in fixed proportions (along the lines of Brunnermeier et al. 2011a, 2011b).1 This would accomplish the goal of preserving market discipline while at the same time solving a crucial unresolved problem of the Eurozone – the excessive holdings of own national debt by national banks.

Our proposal has two complementary aspects, one related to monetary policy and the other to regulatory and supervisory policy.

The monetary policy aspect

  • The ECB would announce that for its QE operations it would use exclusively a bond formed by the senior 60% tranche of a synthetic bond formed of debt of Eurozone countries (in fixed proportions to their GDPs).
  • The ECB would not be involved in the tranching, but would instead simply announce it would use this instrument.

The market would immediately get to work on creating this asset.

Unlike in the Brunnermeier et al. (2011) proposal, no European Debt Agency or any other intermediary need be involved.  Instead, a (small) ECB office would declare senior synthetic bonds as ‘conforming Euro-Safe Bonds’ when they fulfil these criteria, similar to the role of Fannie Mae and Freddie Mac in the US in declaring some mortgages with certain loan to value ratios, ratings etc. as conforming.

The bank regulation aspect

The ECB and the Single Supervisory Mechanism (SSM) would announce that only the senior tranche of the security so produced could be counted as risk-free for the purposes of the risk weighting and liquidity coverage ratio calculations, which implies changing the current treatment of sovereign bonds for these purposes.  Alternatively, the ECB/SSM would impose a risk concentration limit on own sovereign debt and exempt these senior bonds from such limitations.

This second leg of the proposal is not necessary for the success of the first, but it would leverage the impulse of QE to solve some deep structural problems, as we discuss next.

Advantages of the proposal

This proposal would:

  • Reduce substantially the geographic bias in the flight to safety, as the safe asset is (in regulatory terms) a Europe-wide one;
  • Eliminate the moral hazard induced by current arrangements (governments can default in this world, as the banks are protected from the fallout). Markets would thus monitor the governments instead of second guessing the (bailout) intentions of the ECB.
  • Eliminate the diabolic loop between banks and their governments, since a sovereign in trouble does not jeopardise its own banks (who hold a senior tranche of a diversified portfolio).
  • Reduce the geographic segmentation of Eurozone financial markets.

An additional advantage is that either of these solutions would create a large safe asset. As Caballero and Fahri (2014) argue, citing a Barclay’s calculation, the shocks to Italy and Spain together with the drop in AAA-rated asset-backed securities and agency debt has drastically reduced the supply of safe assets, from 36.9% of world GDP to 18.1%. The creation of a large Eurozone-wide security would reverse this trend and go some way towards moving the economy away from the ‘savings glut’ and its distorting consequences.

It is important to emphasise here that this synthetic debt would not involve any risk sharing among different governments or any debt mutualisation. Each government would continue to issue its own debt and face its own interest rates in the market and the junior tranches would reflect default risk.

Transition regime

The transition to this regulatory regime implies a cost. For instance, profits in periphery banks could decrease since they would partially substitute higher yielding bonds for the risk free synthetic. To the extent that these extraordinary profits are the result of excessive concentration, such a drop would be welcome. Moreover, following the ECB’s Comprehensive Assessment Exercises, banks are now well capitalised, and this reduction in income should not make a significant dent in their capital ratios (although it could well affect profitability, as it should, since it derives from excessive concentration). Specifically, the highest sovereign exposure for the very top tier (over $1 trillion in assets) of European banks2 as a percentage of fully phased-in CET1 capital is 105% at Unicredit. Even for this bank, the reduction in net income from our proposal is estimated on the basis of a static scenario to be around €100 million, compared to a CET1 of €39.9 billion, thus easily absorbable by the bank.

Moreover, positive risk weights for sovereign bonds may increase financing costs for vulnerable countries, possibly rekindling solvency concerns in countries with high public debt levels. This risk is countered from the perspective of high debt countries by the existence of QE, which brings about large benefits. On balance, all countries individually benefit from this proposal.

Concluding remarks

With this proposal, QE would not only help the ECB attain its elusive inflation target, but would contribute to the financial stability of the Eurozone and solve some of the euro’s structural ‘birth defects’ by: (1) introducing an instrument for open market operations, the equivalent of the generic bond; and (2) reducing the exposure of banks to their own sovereigns, thus helping to break the link between bank and sovereign risk. Thus our ‘market safe bonds’ proposal facilitates QE and is in turn facilitated by QE.


Brunnermeier, M, L Garicano, P Lane, M Pagano, R Reis, T Santos, D Thesmar, S Van Nieuwerburgh, and D Vayanos, (2011a) “European safe bonds (ESBies)”. 

Brunnermeier, M K, L Garicano, P R Lane, M Pagano, R Reis, T Santos, D Thesmar, S Van Nieuwerburgh, and D Vayanos (2011b), “ESBies: A realistic reform of Europe’s financial architecture”, VoxEU.org, 25 October. 

Caballero, R J and E Fahri (2014), “On the role of asset shortages in secular stagnation”, C Teulings and R Baldwin (eds.), Secular Stagnation: Facts, causes and cures, VoxEU.org eBook.

Draghi, M (2014), “Unemployment in the Euro Area”, Speech at the Annual Central Bank Symposium at Jackson Hole, 22 August. .


1 We thank Jeromin Zettelmeyer for discussions on this proposal. We also thank Alberto Caruso for excellent research assistance.

2 These banks are: BNP Paribas, Credit Agricole, Deutsche Bank, Societe Generale, Santander, BPCE, Unicredit, and ING.



Topics:  Macroeconomic policy Monetary policy

Tags:  Eurozone QE, Safe Market Bonds, ECB, quantitative easing, unconventional monetary policy, diabolic loop, doom loop, sovereign debt, safe assets, savings glut, risk weights, bank capital

Research Fellow with the Productivity and Innovation Programme, Centre for Economic Performance; Professor of Economics and Strategy, Departments of Management and of Economics, London School of Economics; and Research Fellow, CEPR

Chairman and co-founder, Now-Casting Economics; Professor of Economics, London Business School; CEPR Research Fellow; Fellow of the British Academy