After AQR and stress tests – where next for banking in the Eurozone?

Thorsten Beck 10 November 2014



The ECB has concluded and published the results of a year-long painstaking process to go through the books of the 130 largest and most important banks of the Eurozone, adjusting balance sheets and testing the sensitivity of their capital position to two different scenarios, one of which includes a severe economic downturn. This exercise constitutes the entry point to the Single Supervisory Mechanism (SSM), where the ECB has direct supervisory responsibility for most of these 130 banks and indirect responsibility for the rest of the banks in countries that are members of the SSM.

There is a lot to be said and criticised about the asset quality review (AQR) and stress tests, but it certainly constitutes a huge improvement over the three previous exercises in the EU.

  • This assessment was a top-down approach – driven and monitored by the ECB – that focused on creating a level playing field in asset valuation and stress testing across banks in the Eurozone.
  • By combining the asset quality review with the stress tests it increased the transparency and reliability of the stress tests.

However, there are also clear shortcomings:

  • The stress test still does not include the scenario of a sovereign default.
  • It does not include the adverse scenario of deflation, an issue that was less of a concern when the stress test scenario was developed earlier this year.
  • It focuses exclusively on the risk-weighted capital–asset ratio and not on the leverage ratio. Fourteen of the 130 banks before the AQR, and 17 banks after, are below the 3% leverage ratio, and that does not take into account yet the effect of the stress tests.

The results provide interesting insights into the shortcomings of national supervision, as the AQR has revealed glaring discrepancies in asset classification. For example, more than 20% of the reviewed debtors were reclassified as non-performing in Greece, Malta, and Estonia, and even 32% in Slovenia. In the case of one bank, that share even amounted to 43%. This suggests a high degree of regulatory forbearance. These discrepancies between published balance sheets and adjusted numbers presented in the ECB clearly point to the consistency advantages of supranational supervision.

Overall, the comprehensive assessment identified a capital shortfall of €24.6 billion across 25 participating banks after comparing the projected solvency ratios under the adverse stress test scenario against the threshold of 5.5% tier 1 capital relative to risk-weighted assets. Taking into account capital raisings during 2014, only 13 banks still face a net capital shortfall of a total of €9.5 billion, a rather small number. Among the 13 banks, four are in Italy, three in Greece, two in Slovenia, one in Portugal, one in Austria, one in Ireland, and one in Belgium – thus a certain concentration in crisis countries.

The number of banks with (net) capital shortfalls seems exactly what the doctor prescribed – not too low, so that the test would be considered rigorous enough, not too high, in order to not shake the market. As recognised by the ECB itself, this is just a starting point into the SSM; as capital requirements become tighter, additional capital shortfalls might appear. The initial reaction seems to indicate that the ECB has achieved two of the objectives of the exercise: transparency and confidence building. The third objective – repair of banks’ balance sheets where necessary – is still a work in progress.

Where next?

While this seems a good starting point for the first pillar of the banking union, it again lays open the limitations of the second and third pillars. The 13 banks that have net capital shortfalls have nine months to come up with recapitalisation/restructuring plans. While the ECB is involved as supervisor, it is not in itself a resolution authority – a responsibility still at the national level. The Single Resolution Mechanism (SRM) only comes into force in 2016, and even then it is not an effective supranational mechanism like the SSM, but rather the result of a compromise, a mix of national and supranational frameworks. A third pillar, a joint deposit insurance funding scheme has been quietly dropped.

The evolution of the Eurozone crisis over the past five years, on the other hand, provides sufficient arguments for a full-fledged banking union. Just two examples:

  • The example of Cyprus shows that a deposit insurance scheme is only as good as the sovereign backing it. The banking union in its current form does not address this.
  • The recent failure of the Portuguese Banco Espirito Santo shows the limitations of effective resolution in fiscally weak countries. Ultimately, the Portuguese government had to rely on Troika funding to resolve the bank, in light of the precarious fiscal position of the Portuguese government.

Where does this leave us?

There are two ways to look at the current situation, corresponding to the glass-half-full and glass-half-empty attitudes. On the one hand, the fact that some basic elements of a European financial safety net have been put in place is a great success. On the other hand, this compromise is far from the financial safety net that an integrated European banking market would need – and a half-baked banking union might actually backfire, as some observers have pointed out (Schoenmaker 2012).

A lot has been accomplished:

  • A single supervisor that can internalise cross-border externalities. Matching the perimeter of banks with the regulatory perimeter can reduce distortions in the supervisory process. And given the risk of political capture of regulators that we could observe across Europe (both in the core and periphery), this is also progress. Given how politically sensitive banking has been across Europe, this is indeed a big step for the Eurozone.
  • Bank resolution frameworks across Europe are being strengthened, on the national level, but also – with the bail-in clause – on the European level. The SRM – with all the caveats stated below – is an important first step in the right direction towards resolving failing banks in a more effective way than done during the recent crises.

There are still several issues open:

  • What are the relative roles of the European Banking Authority and the ECB in supervising banks in Europe?
  • How will fragilities be addressed during the 14-month transition period between now and the date the SRM takes effect on 1 January 2016?
  • How effective can the ECB be as supervisor, having to deal with 19 different banking laws?

What has been missed:

  • A European bank resolution mechanism that deserves the name. The current arrangement is too complicated and still relies too much on the subsidiarity principle, where first national authorities are in charge and only then the European mechanism comes in place. A single banking market requires a single financial safety net.
  • A proper funding mechanism is missing. While there is some money available, funded by the banking industry itself, it seems too small to serve for the purpose of having to restructure a relatively large bank somewhere in Europe.
  • A public backstop is missing. Even if the resolution fund under the SRM were less limited than it is under current plans, such a fund will never be enough for a systemic banking crisis. A public backstop is necessary.
  • Zero risk weights for government bonds further increase incentives to hold government debt and facilitate the ‘Sarkozy carry trade’.
  • Most importantly, the banking union just agreed on is a forward-looking structure, designed for the next crisis, but not supposed to address the current crisis.

Are we there yet?

There is an increasing sentiment that the Eurozone is about to exit the crisis. With both Ireland and Portugal having exited Troika programs, Greece showing signs of economic recovery, and the crisis in Cyprus being less deep than feared, there are understandable hopes that the worst might be behind us.

And there seems to be an important positive trend across Europe, which is the return to market discipline. The fact that many banks have accessed markets to raise additional equity in 2014, supposedly before being forced to do so by the ECB, can very well be interpreted as the return of market discipline. In the absence of a European banking union and a Eurozone mechanism to recapitalise banks In addition, recent bank failures have seen junior creditors being bailed in rather than bailed out (SNS Reaal and Cyprus), even though the bail-in rule supposedly only kicks in after 2016.

The discussion on the banking union is related to a broader question about the role of the banking system in European finance. As pointed out by many observers, European financial systems are heavily bank-based, with a limited role for non-bank financial providers and capital markets (European Systemic Risk Board 2014). This exacerbates the link between governments and banks, as there are a limited number of non-bank buyers of government bonds. It makes financial systems more concentrated and results not only in larger banks, but also a stronger reliance on large banks.

Concluding remarks

Post AQR/stress tests and post-banking union agreement, the largest risk is not necessarily the half-baked nature of the banking union, but rather the complacency that might follow from the feeling ‘we have done it’. A long-term reform agenda awaits Europe. We are not there yet!


European Systemic Risk Board (2014), “Is Europe Overbanked?”, Advisory Scientific Committee Report 4, June.

Schoenmaker, D (2012), “Banking union: Where we’re going wrong”,, 16 October.




Topics:  Financial markets

Tags:  ECB, eurozone, banking union, stress tests, Asset Quality Review, forbearance, recapitalisation, balance sheets, leverage, bank resolution

Professor of Banking and Finance, Cass Business School; Research Fellow, CEPR


CEPR Policy Research