Response to consultation on Draft Guidelines on the methodology to estimate and apply credit conversion factors under the Capital Requirements Regulation
1. Question 1: How material are the cases for your institution where you would have to assign an SA-CCF to exposures arising from undrawn revolving commitments and thus restrict the use of own estimates of LGDs within the scope of application for IRB-CCF in the CRR3? For which cases would you not have enough data to estimate CCFs but have enough data to es-timate own estimates of LGDs?
/
2. Question 2: Do you have any comments related to guidance on the identification of a relat-ed set of contracts which are connected such that they constitute a facility?
We welcome the proposition from the EBA to leverage on the CRR3 definition of facility, which allows to consider a set of contracts for CCF calculation.
However, we would like the EBA to be more explicit in the Guidelines that LGD and CCF can be calculated at distinct granularities. Indeed, the CRR definitions clarify that both the LGD and CCF parameters are related to a “single facility” as per articles 4.1(55) and 4.1(56), leaving the door open to an unintended reading of regulation that both parameters need to be calculated at the same level of granularity. It is worth highlighting that supervisory reading as per paragraphs 260/316 of the ECB Guide to Internal Models (July 2025) is to still allowing in CRR3 distinct granularities between LGD and CCF. Indeed, the rationale for calculating LGDs at a more aggregated level than the facility level is linked with the recovery processes, while the CCF parameters are linked with the way limits are granted and managed within the bank.
In addition, the restriction of IRB-CCF models to revolving commitments only makes it all the more difficult to align LGD calculation granularity (as LGD models would cover both revolving and non-revolving products) with the CCF calculation granularity.
3. Question 3: Do these GL cover all relevant aspects related to the definition of revolving commitments that you consider relevant for the scope of the IRB-CCF? Have you identified any product that should be in the scope of the IRB-CCF that is currently excluded in the GL? In terms of off-balance sheet exposures, how material are the exposures that fall within the defined scope of the IRB-CCF for your institution?
EBA should nuance its position regarding deferred debit cards. The revolving definition is interpreted with 3 cumulative criteria being met:
(A) The obligor has flexibility on drawings, within a given limit; AND
(B) The obligor has flexibility on repayments: the obligor decides the timing of its reimbursements; AND
(C) The drawing capacity is replenished by the amount reimbursed.
This reading is based on article 166(8b) of CRR: “a commitment shall be deemed “revolving” where it lets an obligor obtain a loan where the obligor has the flexibility to decide how often to withdraw from the loan and at what intervals, allowing the obligor to drawdown, repay and redraw loans advanced to it. Contractual arrangements that allow prepayments and subsequent redraws of those prepayments shall be considered revolving.”
Banks should determine the revolving feature of products based on CRR3. Some deferred debit cards will not meet for instance the criteria (B) / (C) and will not be considered as revolving products. Banks should determine the revolving feature of products based on CRR3/European law definition. In overall, we think that considering increase of EAD for deferred debit cards is less of an issue because their drawings are off-set by the current account balance which is most of time positive.
4. Question 4: Are there products that have an advised limit of zero but a nonzero unadvised limit that should be included in the scope of the IRB-CCF GL? How material are these cases for your institution?
/
5. Question 5: Do you think that dynamic limits (e.g. limits the extent of which is dependent on the market value of financial collateral pledged by the obligor in relation to the revolving loan) warrant a specific treatment in the IRB-CCF GL? How material are these cases for your institution?
/
6. Question 6: Have you identified any unwarranted consequences of including fully drawn revolving commitments in the scope of the IRB-CCF. How material are these cases for your institution?
We do not see in level 1 text any ground for developing an approach for fully drawn revolving commitments which would estimate the EAD above the full on-balance sheet drawn amount. As article 166.8 of CRR indicates "An institution that uses IRB-CCF shall calculate the exposure value for undrawn commitments as the undrawn amount multiplied by IRB-CCF", if the undrawn amount is 0, the exposure value for undrawn part will be 0. In article 182.1c of CRR, only facilities which are close to be fully drawn are concerned by the fact that CCF should be effectively quarantined ("Institutions shall ensure that their IRB-CCF are effectively quarantined from the potential effects of region of instability caused by a facility being close to being fully drawn at the reference date"). We understand that such specific treatment does not include fully drawn commitments.
8. Question 8: Are there cases for your institution where the calibration samples should be shorter than the sample used to calculate the long run average (LRA) CCF
The creation of different samples in the risk differentiation (training versus testing out-of-sample and out-of-time) and in the risk quantification is consistent with the practices (and requirements) of PD and LGD parameters as well, even before CRR3 finalization. Nonetheless, one potential concern is for non-retail CCF models covering low default portfolios : due to low number of observations, it may not be possible to meaningfully split into the required samples.
9. Question 9: Do you have any concerns with the requirements introduced to analyse and mitigate a lack of representativeness for CCF? Do the requirements on the different data samples when observing a lack of representativeness impede your ability to model CCF portfolios?
/
10. Question 10: Do you have any concerns with linking the fixed CCF to the lack of historical data available to the institution in relation to the coverage by the RDS of material subseg-ments of the application portfolio? How is your institution currently treating these cases?
/
11. Question 11: Are there any concerns with requiring consistency in the analysis of changes in the product mix with the institution’s definition of facility? Are institutions able to identify and link contracts (partially) replacing other contracts where the closing or repayment of one contract is related to the origination of a new contract? Are institutions able to link new contracts that are originated after the reference date to related contracts existing at refer-ence date? In particular, is it possible in the case contracts that are revolving commitments are replaced by contracts that are non-revolving commitments (e.g. by a term loan)?
EBA is not consistent in the Guidelines when asking for instance consistency between application and estimation on fully drawn commitments but the selection of perimeter does not ensure such consistency. Indeed, paragraph 60 page 71 of the Guidelines imply notably that if the facility is non-revolving at reference date but it becomes revolving between 12 months before default and default, it is included in the IRBA CCF modelling perimeter. However, in CCF application, at a certain snapshot/reference date of RWA calculation, the bank cannot anticipate futur change of product type for a specific facility and can only observe the product nature at snapshot/reference data. It would be then be inconsistent with the estimation of CCF.
12. Question 12: Do institutions consider it proportionate to the risks of underestimation of CCF to perform the identification analysis and allocation procedure? If it is deemed not propor-tional, what would be an alternative approach that is still compliant with Article 182(1b) CRR?
/
13. Question 13: Do you have any concerns on the proposed approach for the treatment of so-called ‘fast defaults’? In case you already apply a 12-month fixed-horizon approach, do you apply a different treatment for ‘fast defaults’ in practice, (and if so, which one)? Is the ‘fast default’ phenomenon material according to your experience? If yes, for which exposures, exposure classes or types of facilities?
In our view, a variable horizon during the twelve months prior to default is also relevant to capture counterparties’ behavior and it allows to account for changes in facility terms during this period of time.
We believe that the use of a variable time horizon should also be allowed, at least for risk differentiation purposes, where a higher volume of data may be useful, e.g., to assess changes in counterparties' behavior for corporate low-default portfolios
14. Question 14: Do you have any concerns on the multiple default treatment? To what extent are your current models impacted by the application of a multiple default treatment?
/
15. Question 15: Do you agree with the three principles for the calculation for realised CCF in the context of consumer product mix, and their implications for the cases mentioned as ex-amples? In case of disagreement, what is the materiality of the cases with unwarranted re-sults, in particular in relation with the definition of facility applied in your institution? In case of material unwarranted results, can you describe your alternative practice to this CP?
Depending on the products and the representation in data for banks, we suggest that EBA does not impose the allocation methods. It should be up to the banks to ensure consistency between CCF numerator and CCF denominator and detail in procedures the allocation methodologies used in case of product mix or product transformation. Such allocation methods should be in line with the bank's granting process and monitoring of facilities.
Based on the examples introduced in the GL CCF (page 31 to 34), we would like to highlight methodological issues regarding the CCF calculation. In Case I, excluding limit increases from the denominator leads to CCFs inflated to levels with no economic meaning (e.g. 200% instead of 100%), which does not reflect the actual facility usage. In Case IV, only the revolving part of the facility is included in the CCF numerator (EAD), while the same EAD in the LGD denominator of the same facility will captures both term loan and revolving parts, raising concerns on how consistency between CCF and LGD calculations can be ensured, as required by EBA/GL/2017/16. Thus, it seems necessary that banks are allowed explicitely to ensure methodological consistency and to avoid overstating realised CCFs.
16. Question 16: Are there any concerns related to the allocation mechanism described in these GL?
/
17. Question 17: Where credit lines are kept open even if the facility is in default, the alterna-tive option described in this consultation box could lead to high realised CCF values. Is this a relevant element for your institution and if yes, why and how material are these cases with-in the scope of IRB-CCF models?
/
18. Question 18: In case of multiple defaults, the CCF might also be driven by drawings while the obligor was in its default probation period or in the dependence period between the merged defaults. Do you expect this to be material for your CCF models?
/
19. Question 19: Do you see any unwarranted consequences of the proposed approach for in-corporating additional drawings after default? In particular, in order to maintain consistency between the realised CCF calculation and the calculation of the denominator of the realised LGD as described in paragraph 140 of the GL PD and LGD, would this require a redevelop-ment of your LGD models?
The proposition of an unique methodology to include post-default drawings will have material consequences : existing LGD models, that have already been submitted and validated by ECB, will need to be redeveloped only for technical reasons, while the principle of allowing some netting between drawings and repayments makes economic sense.
We suggest to leave more flexibility to the institutions in the choice of the method to be applied, including the possibility to use Max EAD with possible adaptation if any, depending on the characteristics of their portfolios.
In addition, since LGD models can be transversal across revolving and non-revolving exposures, and consistency with the CCF model is required, is it allowed to have two distinct LGD formulas—one for revolving and one for non-revolving exposures—each based on a different approach for additional drawings?
For French banks wishing to apply the EBA regulatory Max EAD formula or any adaptation, the implementation will have an impact on their full scope of LGD models as consistency between LGD and CCF is requested regarding post-default additional drawings. Thus, sufficient time for implementation should be provided in this case.
20. Question 20: Do you think that the relative threshold is an appropriate approach to restrict the use of the alternative CCF approach for those facilities in the region of instability? Do you think it is appropriate to define a single relative threshold per rating system or are there circumstances where multiple relative thresholds would be warranted? Do you see a need to use an absolute threshold in addition to the relative thresholds?
We believe it would be relevant to offer greater flexibility in the approach outlined in the GL, allowing institutions to choose from a variety of options. For example, incorporating both relative thresholds related to CCF and absolute thresholds concerning undrawn amounts could provide additional alternatives to isolate facilities with very low off-balance-sheet exposure and improving the ability of identifying relevant cases more effectively. For the avoidance of doubt, banks can use absolute and/or relative thresholds to identify region of instability.
Besides, we do not define a single relative threshold per rating system. As an illustration, when the volumetry of defaults is sufficient, there could be cases where per model, several sub-regions of instability can be defined ex-ante according to different axis of segmentation. As an illustration, they could for instance concern:
- Transaction / obligor characteristics: different types of products can for instance show different drawing profile which justifies that the region of instability can be distinct between types of products
- Time period / generation: historical data periods can influence the region of instability profile, due to differences in terms of business activities in the older generations
Other aspects could be considered in accordance with portfolio specificities. In some cases, this is only when the ex-ante segmentation in sub-regions of instability is done that we calibrate absolute threshold and/or relative threshold by sub-region of instability.
21. Question 21: Do you consider the guidance sufficiently clear in relation to the requirement for institutions to set up a policy to define a threshold value?
/
22. Question 22: Do you consider it appropriate to set a prescribed level or range for the de-fined threshold, and if so, what would be an appropriate level for the threshold? In case an absolute threshold is warranted, what would be an appropriate prescribed level for an abso-lute threshold?
Given the complexity of the topic, banks should be allowed to calibrate the own thresholds as a one-size-fits-it-all-approach is not possible.
23. Question 23: Do you think that, for the facilities in the region of instability, and/or for fully drawn revolving commitments, a single approach should be prescribed (e.g. one of the ap-proaches above defined in the Basel III framework), or that more flexibility is necessary for institutions to use different approaches they deem most appropriate for these facilities?
We do not think that a single approach is relevant (see answer to Question 24).
24. Question 24: If such flexibility is indeed warranted, what is the technical argumentation why prescribing a single alternative approach for these facilities is not suitable? Which cases or which types of revolving commitments could not be modelled under the approaches pre-scribed? Are there types of revolving commitments that could not be modelled by any of the approaches described in the Basel III framework?
It would be relevant to allow use of other Basel approaches or adaptation of such approaches. As a reminder, Basel approaches consist of:
- Limit Factor approach: the predicted balance at default is expressed as a percentage of the total limit that is available to the obligor under the terms and conditions of a credit facility
- Balance Factor approach: the predicted balance at default is expressed as a percentage of the current balance that has been drawn down under a credit facility
- Additional Utilisation Factor approach: the predicted additional drawings in the lead-up to default are expressed as a percentage of the total limit that is available to the obligor under the terms and conditions of a credit facility
- Each Basel approach may raise the following issues in the calculation:
- The Basel approaches were not written in the context of a 12-month fixed horizon approach. However, one strong assumption is the reference date especially for the denominator in the calculation of each Factor. Such reference date is 12 months before default, thus when applying the Basel approaches, the denominator is the drawn/balance amount or total limit at the reference date. However drawn/balance amounts or total limit amounts could evolve before 12 months before default and the default (thus with a mismatch with amounts in the numerator). Such asymmetry between the numerator and the denominator is structurally an issue in all Basel approaches
- Low total limits at reference date could create extreme values in the calculation under Limit Factor and Additional Utilisation Factor approaches, as the denominator is rather low
- The Balance Factor approach could create instability issues due the denominator being the drawn/balance amount at reference date, which could be disconnected with the drawn/balance amount at default date.
We describe below relevant adaptations of Basel approaches that banks could think of, to illustrate the issues encountered (rather than supporting them as a prescriptive approach):
- In the Additional Utilisation Factor approach for instance, we understand that the calculation of an Additional Utilisation Factor is necessary (being the predicted additional drawings in the lead-up to default calculated as the difference in the drawn amount between default date and reference date, divided by the total limit at reference date). However, to circumvent issues related to low limits, we could express directly the EAD as the drawn amount at reference date + an additional drawing factor. Thus, the EAD will be an equivalent to what appears in the CCF numerator.
- Another way to circumvent the difficulties could be to simplify significantly the approach where the scope is rather limited. In this case, we can think of expressing the EAD as the drawn amount at reference date + X amount. The X amount could be either calibrated or a fixed value.
As there could be different ways to best estimate the calculation depending on the cases as illustrated above, we would favour the EBA providing sufficient flexibility (use of Basel approaches or adaptations) in the calculation approaches in the Region of Instability so that banks can take the most relevant approach for their portfolios.
25. Question 25: Which of the three approaches described in the Basel III framework is pre-ferred in case a single approach would be prescribed?
See answer to Question 24.
26. Question 26: For the purpose of the long run average calculation, are there any situations where such intermediate exposure weighted averaging at obligor level would lead to a dif-ferent outcome (that is unbiased) with regard to the CCF estimation? How material is this for your portfolio?
We welcome the fact that the EBA does not introduce intermediate step of calculation with CCF yearly averages. However we would like to hightlight that this proposal (calculation over the entire historical period, without intermediate steps involving yearly average CCF) is contradictory with the EGIM, which should therefore be updated accordingly, We suggest that in consequence a relevant transitory phase can be applied by banks. This is another example that could disrupt the current model landscape and may require transitional arrangements, to be discussed with the JST on a case-by-case basis.
More generally regarding the calculation of LRA CCF, we propose that the EBA clarifies that basic statistical techniques, such as the treatment of outliers using percentile or absolute caps, may be used to improve the robustness of risk quantification levels.
27. Question 27: Do you have any comments on the condition set to use the simple approach to estimate additional drawings after default. Do you consider that the simple approach is also relevant for retail portfolios?
/
28. Question 28: It was considered that requiring institutions to exclude unresolved cases from the long run average CCF, if their realised CCF is lower than the LRA of the corresponding fa-cility grade, could be seen as too conservative. Do you have any comments on this treat-ment introduced in the simple approach? Do you have specific examples when this treat-ment would not be appropriate?
/
29. Question 29: Do you have any comments on the modelling approach to estimate additional drawings after default for unresolved cases?
In the first place, the EBA approach makes sense from a conceptual perspective as LGD-CCF consistency is a key principle of the IRB framework.
However, implementation of the EBA approach raises operational issues. Indeed, our current LGD and CCF models are currently developed with possible misalignment between LGD and CCF calculation granularities. There would be a first issue of mapping if we were to reuse extrapolation from LGD model line by line. Moreover, as CCF grades are not defined on the same basis that LGD, the mapping for the maximum recovery period could be difficult when it is calibrated by LGD grades for example. One additional difficulty is that the LGD is developed for revolving and non-revolving lines which complexifies the issue. If we were to use extrapolated AD estimates both for CCF and LGD as requested by EBA, this could imply redeveloping LGD models.
30. Question 30: Do you have any concerns with the requirement to use as a maximum drawing period the maximum recovery period set for LGD?
/
31. Question 31: For CCF estimation, do you use estimation methods that incorporate portfolio-level-calibration of the estimates? What are the main reasons to use a calibration at a level that is higher than the grade-level calibration?
/
32. Question 32: Do you have any comments on the guidance for the CCF estimation of default-ed exposures?
CRR lays down a specific RW formula for defaulted assets with the use of LGD-in-default (articles 153.1 and 154.1) which justifies that EBA clarifies LGDD requirements in EBA GL on PD-LGD estimation. When writing the GL on CCF estimation, we understand that the EBA has derived an approach for a CCF-in-default from what is required for LGD-in-default. However, contrary to LGD, regulation does not introduce any distinction in exposure value between defaulted and non-defaulted exposures (article 166). Indeed, the same articles 153.1 and 154.1 of CRR detail RWA for both non-defaulted and defaulted exposures with the same formula: Risk – weighted exposure amount = RW · exposure value. Exposure value is determined by article 166 which does not differentiate CCF between defaulted and non-defaulted assets. There is also no such differentiation in the determination of EL as per article 158.5, with implicit reference to article 166 for the definition of exposure value. Eventually, SA-CCF are not differentiated between defaulted and non-defaulted exposures. Thus, we do not see any ground for the EBA to create a CCF-in-default. This is why our reading is that we comply with CRR when we use performing CCF for all defaulted exposures for retail and non-retail exposures.
As a matter of fact, introducing CCF-in-default models will introduce unwelcome overcomplexity in model landscape with difficulties on some portfolios to retrieve sufficient data to model CCF-in-default. Moreover, there is a lack of proportionality with regards to the concerned IRBA CCF portfolio which is restricted to revolving commitments (with no mandatory IRBF in CRR3.
33. Question 33: Do you have any comments on the determination of the low share of ob-served additional drawings after default in the historical observation period relative to the observed undrawn amount at default date? Do you consider it appropriate to set a pre-scribed threshold to determine what constitutes this low share? If so, what would be an ap-propriate value for such a materiality threshold?
/
34. Question 34: Are there examples where the haircut approach should be considered the most appropriate approach for estimating the downturn CCF?
/
35. Question 35: Do you think the add-on of 15 percentage points is adequately calibrated when the downturn impact cannot be observed nor estimated? Could you provide clear examples or reasons why this add-on should be higher or lower than 15 percentage points?
/
36. Question 36: Have you observed, or do you expect a (statistically significant) correlation be-tween economic indicators and realised CCFs? If so, do you expect higher or lower levels of CCFs observed in the downturn periods compared to the rest of the cycle? Do you have pol-icies in place that restrict or, on the other hand, relax the drawing possibilities in the down-turn periods?
/
37. Question 37: The possibility to have no downturn effect on CCF estimates is restricted to the case where observations are available during a downturn period. Which alternative meth-odologies could be used to prove the non-existence of a downturn effect on CCF estimates, in the case where no observation is available during a downturn period?
As a general comment on downturn estimation, once downturn period is defined for LGD and CCF, the choice of impact years - years used for the calibration of downturn estimates - should be determined conjointly for LGD and CCF. This would avoid downturn impact to be based on both highest LGD and highest CCF without connection to the downturn period, which makes no sense from our point of view