Response to consultation on draft RTS on IRRBB supervisory outlier tests
Go back
While market volatility in 2020 did impact bond yields, as evidenced in the consultation paper, is it truly realistic to further lower the floor for interest rate shocks to ECB policy rates? While the accommodation of bond yields is understood, lowering the floor to cover every eventuality has a material impact both on Banks reported sensitivities (increasing exposure to margin compression) and creates operational challenges.
The application of the revised post shock interest rate floor is not gradual and moderated when applied to the parallel down shock for NII sensitivity, it will be immediately impactful. This could have the impact of doubling Banks' reported NII sensitivity instantly. Add to this the expectation of central bank rate hikes and the NII sensitivity of Banks to a down shock will significantly increase as they move further away from the floor. For example, for a -200bps shock now a Bank may only have c50bps of margin compression before hitting the floor. If the floor is moved to -1.50% then the shock size increases to 100bps. If the ECB deposit rate were to hike by 50bps then the shock size becomes 150bps. The QIS requires Banks to floor at the -1% shock, therefore, this dynamic will not be reflected through the QIS so relying on this dataset to set the threshold of what an NII outlier looks like could result in unforeseen results. An appropriate calibration of the % of Tier 1 capital needs to incorporate the above considerations.
In addition, we do not find the inclusion of market value changes in the NII metric useful.
Metric B tries to apportion administrative expenses in the same proportion as income but this is highly unlikely to mirror reality, especially for universal model banks which have a mix of trading and banking book businesses. Therefore, the ‘alpha’ metric and the assumed proportion of admin expenses assigned to NII is inappropriate as it is an over-simplification and also has a material impact on the outcome of the NII SOT result. A small change in ‘alpha’ can have a large relative impact on NII SOT. In addition to this, the metric does not take into account diversification of income sources between business models, nor does it take into account capital buffers. We do not believe that the size of this metric is a true reflection of the ability of a business to withstand interest rate shocks. In addition, the proportion of operating income attributable to NII may change as rates rise and the historic data used for calibration is from a low rate environment.
Respondents are also kindly requested to express whether they find an inclusion of market value changes in the calculation of the NII SOT clear enough.
The change to the floor will immediately impact Banks where a down shock drives their EVE exposure. As a result, the impact on % of Tier 1 capital could rise and may require a recalibration of the current 15%.While market volatility in 2020 did impact bond yields, as evidenced in the consultation paper, is it truly realistic to further lower the floor for interest rate shocks to ECB policy rates? While the accommodation of bond yields is understood, lowering the floor to cover every eventuality has a material impact both on Banks reported sensitivities (increasing exposure to margin compression) and creates operational challenges.
The application of the revised post shock interest rate floor is not gradual and moderated when applied to the parallel down shock for NII sensitivity, it will be immediately impactful. This could have the impact of doubling Banks' reported NII sensitivity instantly. Add to this the expectation of central bank rate hikes and the NII sensitivity of Banks to a down shock will significantly increase as they move further away from the floor. For example, for a -200bps shock now a Bank may only have c50bps of margin compression before hitting the floor. If the floor is moved to -1.50% then the shock size increases to 100bps. If the ECB deposit rate were to hike by 50bps then the shock size becomes 150bps. The QIS requires Banks to floor at the -1% shock, therefore, this dynamic will not be reflected through the QIS so relying on this dataset to set the threshold of what an NII outlier looks like could result in unforeseen results. An appropriate calibration of the % of Tier 1 capital needs to incorporate the above considerations.
In addition, we do not find the inclusion of market value changes in the NII metric useful.
Question 2: Do respondents have any comment related to these two metrics for the specification and the calibration of the test statistic for the large decline in Article 6 for the purpose of NII SOT? Specifically, do respondents find the inclusion of administrative expenses in metric 2 clear enough? Do respondents have any comment on the example on currency aggregation for metric 1 and metric 2?
Metric A is preferred as the numerator is an existing known metric, namely NII sensitivity. NII sensitivity is actively monitored and managed by banks. Metric A is consistent with the approach taken for EVE SOT where the numerator is a disclosed metric and is equally based on a % of Tier 1 capital. We do not see a strong rationale to deviate from this approach for NII SOT.Metric B tries to apportion administrative expenses in the same proportion as income but this is highly unlikely to mirror reality, especially for universal model banks which have a mix of trading and banking book businesses. Therefore, the ‘alpha’ metric and the assumed proportion of admin expenses assigned to NII is inappropriate as it is an over-simplification and also has a material impact on the outcome of the NII SOT result. A small change in ‘alpha’ can have a large relative impact on NII SOT. In addition to this, the metric does not take into account diversification of income sources between business models, nor does it take into account capital buffers. We do not believe that the size of this metric is a true reflection of the ability of a business to withstand interest rate shocks. In addition, the proportion of operating income attributable to NII may change as rates rise and the historic data used for calibration is from a low rate environment.