An incomplete step towards a Banking Union

Daniel Gros 07 July 2012



Banks were “international in life, but national in death” in the first couple years of the Global Crisis. Large, internationally-engaged banks had to be rescued by their home government country despite the rescue being in the interest of many nations.

How times change. European banks are now “national in life, but European in death”. In Spain, for example, the banks’ problems were all national – local savings banks (cajas) financed a huge real estate boom. As the boom turned to bust and the losses threaten to overwhelm the capacity of the Spanish state, the problem became European. An Irish-like failure of Spain triggered by a collapse of its banks would threaten the euro’s very survival and this core economic interest of all EZ members (De Grauwe 2012).

The Spanish case is symptomatic of a larger problem.

  • National supervisors have always a tendency to minimize problems on their own turf.

In the case of large international banking groups, the instinct (and the bureaucratic self-interest) of the home country supervisor is to defend the ‘national champion’ abroad. But the resistance of the national supervisor to recognise problems at home is even stronger. 

This tendency was very much in operation in Spain. Until very recently, Spanish authorities maintained that the real estate problems were temporary.

  • Spanish banking regulators did not want to own up to the reality that they had, for more than ten years, been engaged in oversight – in both senses of the word.

The tragic sense of the word is that they overlooked the build-up of a huge construction boom whose bust now threatens to bankrupt the entire nation. 

The Irish case was not much different to start with. When the problems started to surface, the Finance Minister first claimed that this would become “the cheapest bank rescue ever”.

Systemic weakness of national bank regulation

Given this predictable tendency of national supervisors not to recognise problems at home, it seemed natural that the cost of cleaning up insolvent banks should also be borne at the national level. At first sight it thus made sense that even in the Eurozone banking supervision remained largely national, with only some loose coordination at the EU level. A ‘European Banking Authority’ was created recently, but it has only very limited powers over national supervisors whose daily work remains guided by national considerations.

This ‘national problem and national responsibility’ notion, however, is not tenable given the vast negative spillovers that would come from systemic bank failures in large EZ nations. The problems might arise at the national level, but they quickly threaten the stability of the entire system.

A nascent banking union

The need to rectify this situation has now finally been recognised by Europe’s leaders who decided at their last summit that the responsibility for banking supervision in the Eurozone should be transferred to the ECB.

They naturally put the ECB in charge given that the integration of the financial system is particularly strong within the Eurozone. Moreover, the ECB is already de facto responsible for the stability of the Eurozone’s banking system. But at present it has to lend massive amounts to banks without being able to judge their solidity because all the detailed information about the health of the banks is still in the hands of national authorities who guard this information jealously. And, as we saw in Ireland and Spain, these are the authorities who tend to overlook the problem until it is too late. 

The ECB already de facto started to assume some supervisory power in a little noticed step when it announced that government guaranteed bank bonds would be accepted as collateral for new lending only if the banks draw up a funding plan which indicates how they will be able to finance their operations without excessive recourse to the ECB.

Creeping dis-integration of the EZ banking sector

Putting the ECB in charge would also help to stop the creeping disintegration process which is not visible in public, but very real. There are several cases of large international banking groups headquartered in countries which today are under financial stress. These groups are being torn apart by the conflicting pressures from national supervisors. 

Consider the case of a bank headquartered in Italy, but with an important subsidiary in Germany. The German operations generate a surplus of funds since Germany saves more than it invests at home.  

  • The Italian parent bank would of course like to use these funds to reinforce the liquidity of the group. 
  • The German supervisory authorities consider Italy at risk and thus oppose any transfer of funds from the German subsidiary to the Italian headquarters.  
  • The supervisor of the home country (Italy) has of course the opposite interest. It would like to see the ‘internal capital’ market operate as much as possible. Here again it makes sense to have the ECB in charge which would be neutral with respect to these opposing interests.

Putting the ECB in charge of banking supervision thus solves one problem. But it creates another one. Can one still hold national authorities responsible for saving banks which they no longer supervise?

This is not a new problem. The De Larosiere Report (2009), which became the basis for the creation of the European Banking Authority (EBA) and the Systemic Risk Board (ESRB), argued that the ECB should not be involved in ‘micro’ supervision mainly because banking rescue and resolution involves tax payer money, which they assumed had to be national.

First comes EZ bank regulation then comes EZ bank rescues

Banking regulation and restitution are difficult to separate – no-one wants to pay for things they cannot control. Economic (and political) logic thus requires that the Eurozone will soon also need a common bank rescue fund. 

Officially this is not fully acknowledged yet, except for a hint in the EZ summit statement of June 28-9 which says that once a system of supervision involving the ECB has been created it would become possible for the permanent rescue fund, the ESM, to inject capital into banks. 

This is how European integration often advances. An incomplete step in one area later requires further integration in related areas. In the past this method has worked well. The EU of today is a result of such a process.  But a financial crisis does not give policymakers the time they used to have to explain things to their electorate. The steps will have to follow each other much more quickly if the euro is to survive in its current form. 

Problems ahead

The worrying thing is that the terrain EZ leaders must cross is heavily mined. Europe does not have the luxury to construct its banking union from a stable situation. This new institution is being set up in the midst of a banking crisis.

There are clearly large losses that have to be realised and allocated.  

  • This means serious distributional conflicts both within and between member countries. 

The most difficult case is going to be Spain. The local savings banks are the weakest part of the Spanish banking system because they specialised in mortgages and lending to developers, i.e. the areas where very large losses are to be expected. A number of these were recently 'privatised', often in the context of mergers. These new institutions then had to raise capital in various forms (shares, preferred shares, subordinated debt).

Given that institutional and especially international investors were not willing to invest in these instruments (not surprising given that the state of the Spanish real estate market), the new capital was raised mainly from domestic investors, often the depositors themselves.[1] 

Who pays for past mistakes?

This leads to the first conflict: Who should bear the losses the (Spanish) investors or the Spanish government? 

As retail investors are also voters, the government (and the management of the cajas) now have incentives to pay back as quickly as possible all instruments that would otherwise be loss absorbing. This seems to be happening on a broad scale. It is thus possible that by the end of this year the weakest banks will have repaid all of their hybrid instruments at par or close to par.  At that point the loss absorption capacity of the Spanish banking sector will be much reduced. 

But this leads to the second distributional conflict: Will the European tax payers want to pay for past losses?  As the answer is presumably 'no', there is thus a danger that by the end of this year it will become impossible to inject European capital into Spanish banks unless either a number of banks have gone into informal insolvency (to bail in other creditors) or the Spanish government has put enough into the system to cover past losses (which it might not be able to do).  The road towards banking union is going to be difficult.

Editor’s note: A shorter version of this commentary was published earlier by Project Syndicate, 2 July 2012.


De Grauwe, Paul (2012). Why the EU summit decisions may destabilise government bond markets,, 2 July.

De Larosiere, Jacques (2009). The high level group on financial supervision in the EU, European Commission.

[1] In the case of Bankia preferred shares were sold only a few months before the extent of the losses became public. This has led to allegations of mis-selling and the opening of a criminal investigation.




Topics:  EU policies

Tags:  Eurozone crisis, EZ crisis, bnaking union

Director of the Centre for European Policy Studies, Brussels


CEPR Policy Research