Brussels, 29 October 2014 – The Comprehensive Assessment of EU banks is a welcome step that should improve confidence in the banking sector. But weaknesses in the assessment highlight the need for a binding leverage cap, said Finance Watch, the independent public interest advocacy group working to make finance serve society.
More than 50% of banks’ credit risk weighted assets were covered by the Asset Quality Review (AQR) and the significant findings of this review should help to clean banks’ balance sheets and boost confidence. The communication aspects of this extensive exercise are also of much importance, given the substantial disclosures involved.
However, weaknesses in the approach of the assessment remind us of the remaining systemic risks and the need for a binding leverage cap.
For example, the calculation of capital shortfalls was linked to risk-based capital requirements, a measure that proved nearly useless in predicting bank resilience in the 2008 financial crisis. As the ECB put it in its report: “(…) the crisis has shown that those requirements alone are not sufficient to prevent institutions from taking on excessive and unsustainable leverage risk”. (ECB report p. 127)
Also, the assumption in the stress tests of a static balance sheet (one that would not change during a period of stress) was a compromise in search for simplicity in conducting the exercise and neglects the second round effects of stress, since banks’ response to stressed conditions and its implications, including for example asset fire sales, are not taken into account. Moreover the decision not to apply new capital rules in full (“fully loaded CRD IV/CRR”) might undermine the transparency and comparability of results. Together, these weaknesses risk creating an illusion of safety around the results of the test.
Paulina Przewoska, Senior Analyst at Finance Watch, said:
“The significant findings of AQR should help banks to clean their balance sheets, a first step to restoring lost confidence in the banking sector. Still, a lot remains to be done to refocus banks on the real economy and to deal with the system’s excessive leverage.”
For more detailed comments on the exercise, please refer to the notes below.
To speak to Paulina or one of the team, please contact:
Greg Ford on +32 2 880 0440 or firstname.lastname@example.org
Charlotte Geiger on +32 2 880 0441 or email@example.com
The qualitative review of trading book processes for banks with trading books exceeding EUR 10 billion showed that 19 out of 26 participating banks had poor or substandard practices in one or more sections of the review. This disclosure should help to improve banks’ practices in this area.
The assessment’s technical weaknesses, among other things, relate to the numerator of the capital ratio used (common equity tier 1) and the denominator (risk weighted assets):
i) Denominator: the last crisis showed that the use of risk weighted assets is not the best predictor of a bank’s resilience in times of stress. As the ECB put it in its report: “(…) the crisis has shown that those requirements (risk based own funds requirements) alone are not sufficient to prevent institutions from taking on excessive and unsustainable leverage risk” (ECB report p. 127). The ECB report found that 17 out of 130 banks had a post-AQR leverage ratio under 3%, for example.
The risk weighted assets calculation relies heavily on bank’s internal models, which have not been verified or validated by the ECB (which will, in its role as single supervisor, review them in future under a Pillar 2 review). Hence, the denominator of the ratio, while already not the best predictor of bank resilience remains also conflicted (bank’s own model), non-transparent and non-comparable between banks.
ii) Secondly, the numerator of the ratio is also not easily comparable among banks from different member states, because the exercise was carried out subject to the phasing-in arrangements for CRD IV/CRR. In other words: there was no full loading of all CRD IV/CRR provisions for all participating banks (the provisions governing deductions from CET1 are especially relevant). The ECB notes in its report that: “discretions with respect to phase-in rules applied at the national level currently constitute a country-specific driver of the magnitude of the transitional adjustments” (p.131).
The impact of using transitional arrangements is significant: as of 1 January 2014, the CET1 impact of total CRD IV/CRR transitional adjustments across all participating banks amounted to €126.2 billion (ECB report p. 130). In other words, the level of CET1 at banks from member states using the phasing-in arrangements was boosted in the exercise by as much as €126.2 billion (to see the results with fully loaded CET1 ratios, see EBA’s website).
iii) The assumption in the stress tests of a static balance sheet (one that would not change during a period of stress) was a compromise in search for simplicity in conducting the exercise but neglects the second round effects of stress, since banks’ response to stressed conditions and its implications are not taken into account.
Detailled explanations about the ECB stress tests given by Paulina Przewoska in the Finance Watch webinar “ECB banks stress tests – why taxpayers are at risk” (24 October 2014).
You can help tip the balance! Strengthen our impact by joining our collective efforts.