Question ID:
Legal Act:
Regulation (EU) No 575/2013 (CRR) as amended
Supervisory reporting - COREP (incl. IP Losses)
COM Delegated or Implementing Acts/RTS/ITS/GLs/Recommendations:
Not applicable
Disclose name of institution / entity:
Type of submitter:
Credit institution
Subject Matter:
COREP Template C 25.00.Validation rule v0641_m.

In 2014, our National Competent Authority has clearly mentioned in the validation of the internal model of KCVA computation, that our institution determines the KCVA using the standard method for the below transactions: - Transactions out of the internal model EEPE perimeter - Transactions which are part of the eligible perimeter but, due to data quality issues, the EAD has been calculated in standard method. This justifies that for a counterparty which has expositions calculated in both standard and internal methods, we will have an advanced and standard CVA charge. In the COREP CVA template, we should state the number of counterparties calculated in advanced method (r020) and in standard method (r030), the sum of all counterparties should appear in r010. Related to eba_v0641_m, we remove the duplicates between the 2 calculation methods to fill the r010, in order to avoid the double counting of counterparties for which we have both of STD & ADV CVA. Knowing the fact that we can have the both methods for a given counterparty, should we report strictly the sum of r020 and r030? Or should we continue to consider the r010 as the sum of counterparties that generate KCVA, disregarding the calculation method used?

Background on the question:

Our question follows a question we received from our National competent authority.

Date of submission:
Published as Final Q&A:
EBA Answer:

In C 25.00 r010 c100, the total number of counterparties for which CVA risk has been calculated should appear. There should be no double counting due to counterparties for which both the advanced method and the standardized method is used. Therefore the severity of v0641_m will be updated from blocking to non-blocking.

Final Q&A