A Litmus Test for Banking Union

Among economists a wide consensus exists that a transfer of supervisory powers to the ECB needs to be accompanied by increased risk sharing at the euro area level, through the establishment of a European resolution mechanism and a joint deposit guarantee scheme. This line of argument is based on the accountability principle, according to which “power and resources should go hand in hand” (Dell’Ariccia et al, 2013). Proponents of this big leap towards banking union hope that it can break the doom-loop between sovereigns and banks.

In the meantime, German enthusiasm for a banking union is fading, as illustrated by Schäuble’s insistence on treaty change. Part of German resistance is based on politicians’ preference to maintain control over banks. But German taxpayers are also rightly wary of writing blank checks for European zombie banks. Grand designs by economists will not take away these Germans concerns. The best way to move policy forward is to identify concrete measures which could persuade Germany that a banking union results in a more stable financial system, at minimum risk to the taxpayer.

The usual suspects are the imposition of higher capital requirements and the introduction of better bail-in regulation. It would also help when the European supervisor would be more independent from national supervisors. These measures have been advocated before by many others and make perfect sense. Yet a different measure would be ideally suited to signal euro area governments’ true commitment to a banking system decoupled from sovereigns, and thereby help to overcome German hesitance. This measure would consist of ending the preferential treatment of governments by banks, by enforcing article 102 of the Treaty of Maastricht.

The sovereign-bank nexus works both ways. While a European resolution mechanism may stop failing banks from dragging down their sovereigns, it does not prevent weak sovereigns from infecting the balance sheet of banks. This problem is well-known and has been a major factor in weakening Greek and Cypriotic banks. It shows little sign of abating, as the Banca D’Italia recently reported record holdings of Italian governments bonds by Italian banks. Sovereigns thus continue to use banks as a major source of funding, especially in times of market stress.

A European resolution mechanism will not change this habit, but does imply that the credit risk of government bonds on banks’ balance sheets will be shared. In other words, a banking union establishes a fiscal union through the backdoor. One can argue about the desirability of fiscal union, but as a by-product of banking union it will lack democratic legitimacy.

The present concentration of sovereign risk in bank portfolios violates article 102 of the Treaty of Maastricht, which prohibits the preferential treatment of governments by banks. Article 102 reads: “any measure, not based on prudential considerations, establishing privileged access by community institutions or bodies, central governments, regional, local or other public authorities, other bodies governed by public law, or public undertakings of Member States to financial institutions shall be prohibited.” Both the standard zero risk weightings of sovereign bonds and their exemption from the EU large exposures directive constitute violations of the prohibition on privileged access. It is hard to argue why non-AAA government debt is exempted from the large exposure directive, when higher-quality private creditors must comply. This is a privilege indeed. It is a sad observation that bank supervisors have condoned these violations since the start of EMU.

Breaking the sovereign-bank nexus requires the enforcement of article 102. This would imply that sovereigns no long automatically carry a zero-risk weighting nor be exempted from the large exposures directive. As a result, the exposure of banks to their individual sovereigns would need to go down to a maximum of 25% of capital. While this measure would require a major adjustment in the distribution of government bonds, the advantages are huge.

  • First, by limiting and diversifying their sovereign exposures banks will become more robust to sovereign shocks. This will increase the stability of the euro-area financial system. It would also reduce the chance that Europe is held hostage by profligate governments, whose negotiating power increases with their ability to destabilize the financial system.
  • Second, by substantially reducing the dependency of sovereigns on bank funding, a conflict of interest can be removed. Politicians may want to be tough on banks from a public policy perspective, but they also like to use their financial lifeline. And it is very hard to bite the hand that feeds you.
  • Third, limiting the easy access to domestic bank funding would strengthen financial market discipline and the incentives for politicians to design and implement credible economic policies.

In the euro area, bank holdings of government loans and bonds are huge (ca. € 2700 bln). Given the scale of the necessary adjustment, enforcing article 102 will thus take time. Governments will need to find alternative buyers for sovereign bonds. This is not a mission impossible. In 2011, the Belgian government showed that it is possible to tap household savings directly through a well-designed bond.

The sovereign-bank nexus needs to be cut both ways. Schäuble should ask his euro area colleagues to give up their easy access to domestic bank funding. This is the litmus test for banking union. If they are not prepared to do this, Germany cannot underwrite euro area banks.



Dell’Ariccia. Goyal, Koeva-Brooks, Tressel, A banking union for the Eurozone, Vox, April 5, 2013.