UniCredit agrees on the appropriateness of the proportionality principle, however it is not fully clear whether materiality and complexity should be the only elements to be taken into account as drivers for proportionality. The nature of the activity and risk profile of the exposures in question as an example are also extremely important elements to consider.
Further specifications are required with respect to smaller entities of a larger group. In particular, the proportionately principle, and therefore a lighter assessment framework, should apply in cases where these entities are not relevant within the banking group in terms of dimension (both absolute and relative to the overall group size).
Unicredit agrees that the validation function should be independent from the credit risk control unit and internal audit, as well as from credit granting functions. We suggest to clarify whether the different organization frameworks described in point b), c) and d) of Article 10 are considered as equal alternatives or, on the other hand, an increasing demand for separation is required consistently with complexity/size of the Institution or according to the proportionality principle described in Article 1).
The RTS requires the time series length considered to calculate the long-run average of one year default rates to be representative of an entire economic cycle. More specific instructions should be provided in terms of definition of economic cycle and length of the underlying time series. This would grant consistency across Banks and supervisory practices.
We agree on the on prescriptions related to Article 51. No impacts are expected.
We agree on the on prescriptions related to Article 52.
Article 60 is not completely clear with particular reference to the text box. The meaning of “funded guarantee” should be elaborated as according to our views the guarantees are unfunded by definition.
An accurate evaluation of costs will be possible only after some clarifications and release of the final RTS. In this respect the expected implementation timeline is essential.
The RTS can support the harmonisation of supervisory practice. However this can be reached only if rooms for interpretation are minimized.
We do not expect that these draft RTS will trigger material changes to the rating systems. However firm conclusions can be driven only after some clarifications and release of the final RTS.