- We welcome the guidance on the circumstances where an institution shall be considered as failing or likely to fail. The guidance is clear and would allow sufficient flexibility and discretion to resolution authorities and/or competent authorities to take appropriate decisions in times of stress.
- We believe that the inability of an institution to fulfil its obligations due to an IT problem should not lead to resolution, whatever the duration or impact of the problem: a resolution action would not solve the issue and may create risks of its own.
- We do not believe it would add value to specify a quantitative threshold regarding decreases in asset values across institutions. Other, institution-specific elements (including the availability of recovery measures) could play a major role in such a situation.
We have chosen to respond only to the questions that were most relevant to us.
Example b) could be further clarified. If the inability to fulfil its obligations is related to a (temporary) IT problem and this can be ascertained, a resolution action would not be an appropriate solution, even if the institution appears unable to satisfy its obligations.
It is not clear to us whether you consider the criterion for determining whether resolution should be activated be the size of related obligations (“the obligations that cannot be met are relatively small”) or the duration of the inability to pay (“because it can solve the problem in a timely and effective way”). The size of obligations is independent of the failure to meet them, it should thus not be considered as a valid criterion – even though it may be a concern from a macroprudential point of view and may warrant other regulatory actions. The expected speed at which the problem would be solved is of importance, but is by definition unknown ex ante.
We believe that resolution should only be considered in situations where the institution is heading towards insolvency if no action is undertaken, and where a resolution process may be of help. This is not at all the case for an IT outage. In such a situation, appropriate communication (and possibly temporary liquidity relief) would be sufficient to prevent failure.
In practice, if this were to occur in a situation in which overall market confidence was not affected (no systemic crisis) and in which there were no doubts about the solvency of the stricken bank, that bank would be able to either borrow from the markets or agree with its creditors to pay later (which boils down to the same thing). Only if the outage occurred at a time at which markets were closed could the former option be unavailable.
At any moment in time, it is possible to determine a value of assets that would be lower than liabilities hence define as from which change in asset value an institution would be at risk. However, this (change in value) may be very volatile, which make it difficult to define a static level to be monitored.
In addition, whether the institution would be truly at risk may also, in the case of institutions that are part of a group, depend on the potential availability of intragroup support or, more generally, on the available recovery options.
We agree that all elements in the list are relevant, but we believe that grouping them under the term “operational capacity” is misleading. The first bullet point relates to liquidity risk mainly, the second may be related to operational access to payment, clearing and settlement systems (though it is unclear whether that is indeed your intention) and the third to operational (reputational) risk.