Question ID:
2020_5368
Legal Act:
Directive 2013/36/EU (CRD)
Topic:
Supervisory reporting - Supervisory Benchmarking
Article:
78
COM Delegated or Implementing Acts/RTS/ITS/GLs/Recommendations:
Draft ITS on Supervisory Reporting of Institutions (for benchmarking the internal approaches)
Article/Paragraph:
Annex IV, Template 101, 102, 103, column 100
Disclose name of institution / entity:
No
Type of submitter:
Credit institution
Subject Matter:
Weighted average of the CCF
Question:

Internal authorized EAD models may calculate the EAD by means of several parameters, since the exposure behavior could be explained by different factors.
Considering that this type of EAD internal model does not estimate a unique CCF, but rather several factors that are applied to different exposure components, the consequent approach implemented to report the CCF in Template C.101 ,C102 and C.103 has been the following: include in the field as per C.100 in Annex IV the parameter that, in the internal EAD formula, is applied to the undrawn amount, in line with the weights that shall be used to compute the average of the CCF at counterparty/portfolio level, according to the provision reported in Annex IV which make reference to art. 166, par 8 of CRR.
Please confirm that this approach is in line with the provision set in ITS as for Annex IV C.100.

Background on the question:

The ITS Instructions of EBA’s Supervisory Benchmarking, as in Annex IV for field C.100, requires that, in order to compute the average CCF at counterparty/portfolio level (i.e. Template C101, C102 and C103), the weights that shall be used shall be the amounts to which the CCFs are applied in order to obtain the EAD.
Internal EAD models that foresee multiple factors for the computation of the EAD are applied to different amounts related to the same facility, for instance undrawn amount and granted amount.
The following example shows a possible calculation of EAD based on authorized Internal models which foresees multiple parameters:
EAD=(p_1*undrawn+ p_2*granted+ p_3*DrawingsAfterDefault+ ...)*add_on
As set in Annex IV C.100, in line with the Article 166 (8) of the Regulation EU No 575/2013 (CRR). which specifies the undrawn amount as the basis for the EAD computation in case of above described Internal authorized Model, it has considered the sole undrawn amount: therefore the parameter reported as CCF (C.100) is the factor of the EAD formula which is applied to the undrawn amount, i.e. p_1 in the example formula above, weighted to the related portion of undrawn amount (to which p_1 is applied).
This approach would reflect the EAD internal modelling methodology, strictly based on model components, thus not including possible add-ons (which could be either regulatory add-ons or explicit margins of conservatism) applied on the top of the EAD formula.

Date of submission:
15/07/2020
Published as Final Q&A:
27/11/2020
Final Answer:

In accordance with the instructions for column c0100 of templates C101, C102, and C103 provided in Annex 4 of Regulation (EU) 2016/2070 (ITS on Supervisory Benchmarking), institutions shall calculate and report the weighted average of Credit Conversion Factors (CCFs). The weights to be used shall be the amounts, to which the CCFs are applied in order to obtain the EAD.

Based on the reference to Article 166(8) given in Annex 4 and with reference to Article 4(1)(56) of Regulation (EU) No 575/2013, `CCF' means the ratio of the currently undrawn amount of a commitment that could be drawn and that would therefore be outstanding at default to the currently undrawn amount of the commitment. Also with reference to Article 166(8) of Regulation (EU) No 575/2013, the relevant amounts are those of the committed but undrawn facilities mentioned therein.

Therefore, the CCF values that shall be reported for the purpose of the ITS on benchmarking of internal approaches shall be derived from the final EAD values as used in the calculation of the RWA and not from particular components of the estimation model. 

Status:
Final Q&A
Answer prepared by:
Answer prepared by the EBA.
Image CAPTCHA