The EBA’s stress test 2016 has reached its peak. Institutions submitted to the EBA the results and explanatory note for the first part of the test in April. Now, it’s time for the supervisors to run their process of challenging the banks’ submissions.
In such an exercise, the essential question is one of the hurdle rate – the capital position an institution has to hold after stress effect to be considered as sound. In the 2014 exercise this was 8 percent CET1 for the baseline and 5.5 percent CET1 for the adverse scenario. Unlike the 2014 stress test, the 2016 test has no official or publicly communicated hurdle rate to pass. This sounds like good news at first glance, but includes some uncertainty, however.
One approach to determine what the hurdle rate may be is to consider the minimum requirements. For example:
- For banks under direct supervision of the ECB the SREP-requirement is (on average) 10 percent CET1, which includes the CRR-minimum of 4.5 percent and a Pillar 2 add-on.
- Most, if not all, banks under direct supervision of the ECB have to fulfill a buffer-requirement for domestic or global systemic importance (Basel G-SIB or D-SIB Buffer); for illustrative purposes, let’s assume 1 percent.
- Many banks have to comply with a buffer that should counterbalance systemic risk. Let’s assume this one is set at 0.5 percent.
In this case, the minimum CET1-ratio for the bank in the example above would be 11.5 percent (i.e. 10+1+0.5) accordingly.
So, what is the hurdle rate to pass the stress test in this example? In the baseline scenario, it should be simply the above mentioned minimum requirement of 11.5 percent CET1. Things get a bit trickier in the adverse scenario, however. This is even more important as experience from 2014 shows that most banks struggled more to pass the mark in the adverse scenario. Three scenarios of uncertainty, can be considered as shown below:
- It is not transparent whether the so called “capital conservation buffer” (CCP) is fully front-loaded in the SREP requirement of the ECB. By law, the full buffer of 2.5 percent CET1 is phased in over 4 years from 2016 (i.e. 2.5 divided by 4 which equals 0.625) until 2019 (2.5 percent CET1). The purpose of this buffer is to counterbalance the stress effect. So, it would be reasonable to use this in times of stress. If one argues that the ECB has front-loaded it fully with the SREP 2015 this would imply that the hurdle rate for the adverse scenario would be 250 BP lower (compared to the one in the baseline scenario), i.e. 11.5 percent minus 2.5 percent equals 9 percent CET1 for all years of the stress test (2016, 2017, 2018). If one sticks to the phase-in of the buffer, the hurdle rate would be reduced by 0.625 for the stress year 2016, 1.25 for 2017 and 1.875 for the year 2018.
- It is unclear whether the ECB requires a full capitalization of the stress effect or may use its discretionary power.1 From a statement made by Mario Draghi, ECB President, in a press conference on 10 March 2016, it became clear that supervisors have a great deal of discretionary power. The question is then, how the Supervisor is going to use it. In this context, it is likely that the governance of the stress test will play a primary role. ECB officials made clear several times that the governance – i.e., how the stress test was handled, timeliness of delivery, severity of the findings in the Quality Assurance – is of equal importance as the quantitative results. One could also argue that the “perceived risks” from other parts of the SREP play a major role as well. For example, a perceived weak business model could lead to the requirement to capitalize more stress effects to be on the safe side.
- Finally, we consider the use of other buffers (systemic buffer, D-SIB Buffer, etc.) in times of stress. As the purpose of such buffers is not to counterbalance stress effects, we think it’s highly unlikely that these buffers can be used.
In summary, the hurdle rate(s) for the bank described above can be considered to be:
- 2016 between: 9 percent CET1 and 10.875 percent CET1
- 2017 between: 9 percent CET1 and 10.25 percent CET1
- 2018 between: 9 percent CET1 and 9.625 CET1.
Depending on the individual “performance” of the bank in the stress test and the performance in the overall SREP 2016 process, the individual hurdle rate is skewed more to the left or right of that interval. Given the sheer amount of capital requirements, a proper performance in both exercises – SREP as well as stress test – is key to ensure a proper return on capital. Experience shows that both exercises are not easy: insufficient data quality and automatization, manual processes, ad-hoc developed tools and solutions, tight deadlines and uncertainty of requirements pose challenges.