« Back

French banking federation

Question 1: Are you aware of any situations where the existence of a control relationship among clients does not lead to a ‘single risk’?
Yes. It is typically the case in jurisdictions where banking structural reforms were adopted. In France, the law for regulation and segregation of banking activities provides that trading entities must be segregated and subject to individual capital and liquidity requirements, while the parent company has to apply the Large Exposure regime to its trading subsidiary. Albeit there is still a control relationship between these entities, there is no economic risk anymore and no existence of a “single risk”.

Moreover this might be the case in few situations, in particular for major international Groups acting among several business lines or for specialised lending involving non-recourse debts; even if a control relationship exists, a specific analysis can demonstrate that no single risk exists.

For FBF members, it is important that the internal analysis according to experts can still be considered in the revised guidelines, and that internal risk management practices do not have to be reconsidered.
Question 2: What is the likely impact of the clarification of having an exceptional case when the existence of a control relationship does not lead to a ‘single risk’? Please provide an estimation of the associated quantitative costs.
FBF members see no significant impact expected as their internal procedures already specify the need of specific analysis and documentation of the conclusion in such cases.
Question 3: Do you see a need for further clarification of the accounting provisions which are relevant for large exposures purposes? If yes, please point out the exact indicator of control according to the Directive 2013/34/EU or Regulation (EC) No 1606/2002 which should be clarified with respect to the large exposures regime.
Question 4: Are there any other indicators of control in the case of a similar relationship which are useful to add to this list of indicators?
Question 5: What would be the cost of the assessment of the existence of control relationships in the case of subsidiaries exempted from accounting consolidation? Please provide an estimation of quantitative costs. In your experience, how significant are these cases?
We do not estimate any significant additional costs as our internal procedures to identify individual customers for clustering purposes already require such assessment and documentation (in particular for significant informal groups).
Question 6: Is the guidance provided in section 5. ‘Alternative approach for exposures to central governments’ clear? If not, please provide concrete suggestions.
Yes. However we are very much concerned by the potential impact that such an alternative approach would have on the bank’s sovereign exposures amounts should this be applied. The “assimilated” Sovereign exposure definition is neither defined at the EU level nor at the international level. The FBF is not in favour of such an approach and ask the EBA to wait for new insights at the international level as regards sovereign risk.
Question 7: What is the likely impact of considering that clients are connected as soon as the failure of a client would lead to ‘repayment difficulties’ of another client? Please provide an estimation of any associated quantitative costs.
Banks in general will have to face a major problem of feasibility and we fear that, as a result, the clustering will not be homogenously implemented among banks. Moreover, banks will face the risk of misguided assessments over some groups, leading to non-desirable impacts over the delegation levels and the limit consumptions. Therefore, in order to ensure the feasibility, coherence and comparability among European financial institutions, we suggest that either BCE/EBA or NCAs provide the list of such customer connections.
In addition, we are using this opportunity to draw attention to the need of expending the use of the global Legal Entity Identifier (LEI) to all financial markets (and not limiting its application to specific market segments). We consider such improvement as a major prerequisite for ensuring the consistency of the Large Exposures framework implementation within Europe. In this perspective, we very much welcome the opinion drawn by the ECB on the 12 September 2016 on the proposal for a Regulation amending Regulation (EU) No 345/2013 and Regulation (EU) No 346/2013, standing that “ECB strongly supports the use of internationally agreed standards, such as the International Securities Identification Number (ISIN) and the global Legal Entity Identifier (LEI), as unique identifiers to meet reporting requirements on the securities markets. (…) The ECB is accordingly of the view that, where appropriate and to the extent possible, other legislative changes underpinning the CMU should also establish the mandatory reporting of unique identifiers”.
Question 8: Are the situations described in the list in paragraph 23 as constituting economic dependency clear? If not, provide concrete suggestions. In particular, do you have any comments regarding the introduction of the threshold of ‘at least 50%’ in points c), d), f) and g)?
In the assessments of situations where economic dependencies may exist, it is indicated that the institutions should automatically cluster customers together if one of criteria is met without taking into account the expert analysis. The proposed mechanism does not seem right as, firstly, they are not sufficient on a stand-alone basis to systematically justify a grouping, and secondly, such a systematic process will not consider the dynamic aspect of the economic dependencies. Therefore, automatic criteria will produce misleading results. For example, in point h), the situation where two companies have the same customer base don’t systematically lead to a connection because if one company lose a client, this same client could be retrieve by the other company due to the limited possibility to find new customer. In this case, the expert analysis could be crucial to decide if the connection is relevant. Consequently, we believe that criteria c/, d/ and h/ may be proposed as an element of the assessment but not as systematic triggers of connection.

In our opinion the guidelines should be modified to indicate that the listed criteria should be considered in the assessment process, but without any systematic grouping.
We propose to amend the guidelines par. 23 in the following way:
Replace “institutions should deem, […]” by “institutions may consider, in particular […]”. This will leave the flexibility for banks to assess whether the criteria characterises a real situation of economic dependency or not.
Question 9: Are you aware of any other situations that should be added to the list of situations that constitute economic dependency? In relation to the situation described above, would you treat these exposures as connected? Please explain.
No. We remind that these criteria should not lead to automatic grouping of clients. An internal expert analysis should remain the basis for assessing such connectedness.
Question 10: Is the guidance in section 7. ‘Relation between interconnectedness through control and interconnectedness through economic dependency’ clear? If not, please provide concrete suggestions.What is the likely impact of this guidance? Please provide an estimation of the associated quantitative costs.
Yes. However we consider it regrettable that this consultation is performed without being linked to the on-going discussions relative to the default contagion rules.
Contact name
Sarah Quemon