VoxEU Column EU institutions Europe's nations and regions

Banking union in Europe and other reforms

With most of the debate around banking union in the Eurozone focusing primarily on the financial institutions it will regulate, this column argues that the issue of sovereign debt of the members of the Eurozone needs also to be taken into account.

Many observers have rightly noted that problems in the banking system have been at the core of the ongoing European crisis. In response, policymakers in Europe have proposed a banking union. In particular, the European Commission has proposed that the ECB have broad authority over all banks within the supervisory mechanism (Veron 2012). While other articles in this Vox eBook will focus on additional aspects of banking union that are critical to its success (such as adequate and centralised resolution authority for unwinding of banks), I want to highlight in this note the importance of dealing with sovereign debts of member countries in the Eurozone.

The European banking system is heavily entangled with the debt of its member countries, and the bank exposure to troubled countries (Greece, Italy, Portugal, Spain and Ireland) has led to a loss of confidence in the banking system. Such entanglement, on the one hand, creates credibility for banks and other creditors of sovereigns that other countries and the IMF may come readily to the rescue in case of increase in sovereign credit risk. On the other hand, if sovereign credit risk increases to a point whereby political will elsewhere to support the sovereign becomes doubtful, it leads to severe downward revision concerning the health of the banking system. Ex ante, the banking system may be prepared to take this downside risk, especially if it is highly undercapitalised to start with (so that exposure to troubled sovereigns becomes an attractive version of a ‘carry trade’), or if domestic banks wish to ensure their failure is in a systemic state (downside risk for the entire economy) rather than an idiosyncratic one1.

While a banking union and credible resolution authority would mitigate one side of this problem – that undercapitalised banks engage in seeking sovereign credit risk – if sovereigns are keen to borrow more, e.g., to continue with fiscal excesses or covering-up of underlying lack of productivity in the private sector with an expansion of the government balance sheet, then the entanglement may readily arise again in future. Sovereigns could resist revisiting the zero risk-weights for their debt in bank capital requirements, making it attractive for banks to hold their debt as a way of enhancing levered equity return. Liquidity requirements which are aimed as a prudential tool for banking stability may, somewhat perversely, become an easy mechanism for channeling deposits to fund government deficits. The pressure on central banks to lend reserves against all sovereign debt as collateral can attach liquidity properties to sovereign debt that make its yield low even for sovereigns on path of unsustainable deficits.

Hence a fuller solution to the problem of entanglement of sovereign and banking sectors requires not just a banking union in Europe but direct addressing of the sovereign excess in the borrowing markets. Some reforms that may help facilitate this are straightforward and should be brought to the table in policy discussions of the European Commission, the ECB, the European Banking Authority, and other relevant bodies:

  • Sovereign debt risk weights should be adequately sensitive to the risk of the underlying sovereign, and not be allowed to ‘race to the bottom’ due to lack of participation of some countries in such a revision. Some sanctions in the form of restricted single-market access for banks of sovereigns that do not participate in a revised sovereign risk-weight scheme may be necessary.
  • Liquidity requirements in bank regulation should similarly not treat all sovereign debt holdings as identical regardless of the sovereign’s credit risk. The eligible liquidity holdings must be in the highest quality bucket and possibly diversified across sovereigns whose debt qualifies for this bucket.
  • The ECB should make the haircuts in taking sovereign debt as collateral in line with the sovereign’s credit risk, and in fact, require minimum solvency criteria for sovereigns for their debt to qualify as eligible collateral. To the extent that such haircuts are likely to come under duress in a crisis situation, a rule-based approach with discretion primarily to revise downward the eligibility of risky sovereign debt may be needed.

All of the approaches above suffer from the problem that sovereign credit risk may alter swiftly as it did in 2008-09 due to revelation of hitherto unknown debts (e.g. Greece), extensions of blanket bailouts to banks (e.g. Ireland), reluctance to undertake adequate fiscal cuts, and moral suasion of the financial sector ('financial repression') to hold sovereign debt. In such cases, at least for reasonably large sovereigns, it may end up as incredible to not support the banks that are holding substantial quantities of sovereign debt. In turn, anticipation of this would allow the sovereign ex ante to continue building up of debt. Hence, the Eurozone may need another approach that breaks the sovereign-bank entanglement more directly:

  • An attractive proposal to achieve this has been provided by the Bruegel think tank. One interpretation of this proposal2 is that a central debt management office in the Eurozone will decide how many ‘blue’ bonds can be issued by a sovereign, for instance, based on the ability of the office to match the sovereign’s blue bond issuance against fiscal transfer from the sovereign to the office. The blue bonds can be held by banks; the idea is that these bonds are effectively collateralised and the issue of ex-post bailouts with resources other than those of the sovereign does not arise. However, if the sovereign wishes to borrow beyond the blue bond limit, then it must issue ‘red’ bonds. The red bonds cannot be held by banks, or in other words, they must be held by alternative investors such as hedge funds, pension funds, insurance companies, or more generally, those institutions that are not systemically important. This way, sovereign debt that is not effectively collateralised by tax receipts of the issuing sovereign can be credibly made to bear losses. In anticipation, the market for such debt would effectively discipline the sovereign from over-borrowing by reflecting its credit risk in bond yields and by doing so in a timely manner.

To summarise, banking union in Europe, as and when it is fully achieved, will likely manage the flow of credit risk emanating from weak banks to the balance sheet of their sovereigns. However, it is equally important to manage the flow of credit risk emanating from sovereigns to the banking system holding sovereign debt. To achieve the latter, some explicit steps need to be taken to ensure adequately risk-sensitive capital and liquidity treatment of risky sovereign debt, as well as to directly limit the ability of sovereigns to entangle banking system with their debt without advance collateralisation of such debt.


Acharya, V, I Drechsler and P Schnabl (2011), “A Pyrrhic Victory? Bank Bailouts and Sovereign Credit Risk”, working paper, NYU-Stern.

Acharya, V and R Rajan (2011) “Sovereign Debt, Government Myopia and the Financial Sector”, working paper, NYU-Stern.

Acharya, V and S Steffen (2012), “The Greatest Carry Trade Ever: Understanding Eurozone Bank Risks”, working paper, NYU Stern.

Delpla, J and J Von Weizsacker (2010), “Eurobonds: The Blue Bond Concept and Its Implications”, Bruegel Policy Brief, 2010/03, May.

Veron, N (2012), “Europe’s Single Supervisory Mechanism and the Long Journey Towards Banking Union”, Bruegel Policy Contribution, Issue 2012/16.

1 For the role of entanglement of sovereign debt with banks as a commitment device for the sovereign to repay non-bank creditors, see Acharya and Rajan 2011. For empirical evidence consistent with bank exposures in the Eurozone to GIPSIs being a form of “carry trade”, see Acharya and Steffen 2012. And, for theoretical and empirical discussion of the two-way feedback between banking sector and sovereign credit risks, see Acharya et al 2011.

2 See Delpla and Weizsacker 2010.

945 Reads