Response to consultation Paper on draft RTS and ITS on benchmarking portfolios
Go back
Would competent authorities be expected to investigate any “outlier” contribution and provide the EBA with a feedback, then we advocate the proposed benchmarks are overly prescriptive and disproportionate. For the sake of illustration, the “1st and 4th quartile” rule would systematically require competent authorities to enquire on half of the contributions. Similarly, the “80% SA output criteria” is likely to result in most contributions being considered as outlier for market risk test portfolios with diversification/netting features.
Alternative proposals are discussed in answers to questions 3 and 4.
We recognize the importance of using common and objective benchmarks to identify extreme values. However we consider the current criteria (1st and 4th quartile irrespective of the dispersion of the contributions) is unsuitable.
Starting from the EBA’s proposal an alternative approach could be:
Firstly in order not to get too many outlier banks, to consider other brackets quantiles: e.g. 1st and 9th deciles 0% and 90% instead of the 1st and 4th quartiles.
Secondly, among those outliers banks, to filter only those which absolute differences to the consensus (mean or rather median) are more important than a given threshold, which could be: a Euro amount, a percentage of a consensus on the portfolios mark-to-markets, a number of standard deviations (of the results distribution), etc.
We believe that selecting 50% of the sample (i.e. 1st and 4th quartiles) will create additional works for supervisors, EBA and banks with limited added value. We think outliers should be defined as banks ticking several criteria on the same portfolio, instead of the reunion of banks ticking at least one of the criteria.
At a minimum, the number of outlier contributions should depend on an indicator of dispersion. We recommend using the following formula:
n=max{F ,N ∙(α∙σ/Avg+β)}
Where:
n is such that the n smallest and n highest values are deemed outliers
N is the total number of contributions
σ is the standard deviation of the N contributions
Avg is the average of the N contributions
α and β are scalars to be calibrated
F is a floor level to be calibrated
Note that the above formula can be refined to account for asymmetry in the distance to average of above-average contributions and the distance to average of below-average contributions.
In addition, in case of bimodal distributions (which may happen for instance on some LGD models), the abrupt consideration of quartiles may prove inefficient. For instance, if a portfolio may result either in a high or low LGD with equal probability, Q1 and Q2 would correspond to the same value (high LGD), and so would Q3 and Q4 (low LGD).
We also fear that an inappropriate use of quartiles may result in the herd behavior the EBA wants to avoid: we can indeed assume there will be a strong pressure on banks to exit Q1 (assuming Q1 is the least conservative). On the other side, a position in Q4 is unlikely to result in decreases in overly conservative risk parameters (would banks positioned in Q4 be invited to renegotiate their margin of prudence or VaR multiplier…?).
Comparison between own funds under the internal models and the standardized approach
As developed in answer to question Q1, we reiterate our view that defining a common reference level of own funds at test portfolio level is inappropriate.
The appropriate level of capital for a given portfolio is firm-specific and must be assessed in the light of the firm’s ability to risk-manage such a portfolio, the internal model performance and the representativeness of the portfolio with respect to the overall internal model.
Regarding SA metrics, on top of being rough and non risk-sensitive there is not always unicity in its implementation: for instance on the credit risk side, see the diversity of national supervisory adjustments, or the treatment of sovereign risk under the permanent partial exemption in some countries. On market risk, within the course of SIGTB market risk benchmarking exercise Phase 2, CRM floor contributions were even more volatile than CRM contributions themselves. As a result, it is expected that SA contributions will be highly variable.
As far as market risk is concerned, should SA metrics be ultimately maintained as a common benchmark, we urge the EBA to consider at a minimum the following:
The application of SA output as a benchmark should be postponed until the FRTB new SA approach is implemented,
The computation of SA metrics should be performed at EBA level and the methodology used made available to all participating banks.
On the credit risk side, we are also unsure whether the EBA actually makes a direct reference to the standard approach, since our understanding of Article 4.1.a is that for the purpose of Art.78, sovereign exposures should be risk-weighted on the basis of the counterparty rating, thus disregarding potential existing local specific regulatory treatments - such as the exemption enabling banks to use a nil RW on a large sample of sovereign risks. Should this reading prove correct, we think this not only modifies the computation of Standard Approach as currently implemented by banks, but also opens the way to a new variety of alternative Standard computations, which may be required by various regulatory texts in the future: this would prove extremely complex, and would prevent from a direct comparison between the figures banks are used to reporting, and what they would then have to compute under such alternative Standard approaches. The additional operational risk arising from such manipulations has to be considered as well.
Comparison between estimates and outturns
EBA should give precise guidance on the computation of outturns (such as a spreadsheet), and these computations should take into account all components of the statistical validity of risk estimates, such as the number of years of available history, the default sample size, the number of models in use on the concerned cluster…
For example let us imagine a portfolio with decreasing origination over the recent years, because of careful risk management considerations: the cost of risk on such a portfolio is likely to be on the rise over a 5-year history, however the relative information brought by the 5 annual default rates is not identical due to the change in risk profile and risk appetite. How should such changes be reflected in the confidence interval?
Besides, according to us, the 97.5% level of confidence is a completely new paradigm which may prove inappropriate on some portfolios (owing to various sample sizes, or to the weight of expert judgment in the rating process) and extremely difficult and burdensome for banks to implement. We are not aware of any economic or academic studies proving the importance or relevance of such a confidence level and, therefore, we question the reasons why the EBA is putting forward this new standard. We strongly recommend that the calculation of RWA* and RWA** be abandoned due the many technical difficulties for banks to implement it and for competent authorities to interpret it.
Market risk
Our proposal consists in submitting market risk test portfolios to the back-testing test currently applied at trading book level. This is consistent with the Basel FRTB approach which introduces back testing at a much more granular level.
The flipside of the coin is that back-testing implies that the observation period is much longer than currently proposed. One possible option would be to make the exercise continuous over time while setting up clear rules for rolling tests periodically on underlying instruments of selected portfolios.
Credit risk
We understand the assessment of internal models is one of the objectives of the “General provisions” laid out in art. 8 of the RTS regarding the assessment of the credit risk internal models. Nevertheless we do not figure out accurately to what extent the competent authorities will make use of the information quoted in art. 8-1 and 8-2.
Besides, to avoid creating false outliers, Expected losses (%) or RW (%) should form the first level of comparison. Otherwise, two banks, one with a very early default definition and the other with a very late one, will tend to have high PD/low LGD and (respectively) low PD/high LGD whereas the results in EL% or RW% might look very similar.
For the initial 2014 initial exercise, we highly recommend the use of the BCBS test portfolios. Indeed, we learnt from previous market risk benchmarking exercises that proper portfolio specification, consistent booking across participants and pre-validation exercise are key steps to ensure the exercise meets its target.
Given:
• Other ongoing regulatory exercises with colliding timelines (FRTB QIS, BCBS/EBA joint counterparty risk and CVA benchmarking exercise to mention only 2),
• The very tight timeframe and close kick-off date contemplated by the EBA for the 2014 exercise,
We consider conditions are not met to ensure the 2014 exercise can be properly performed on the basis of a new set of portfolios.
As described earlier, we favor the use of Basel portfolios for the first exercise to take place in Q4 2014 given that:
• The portfolios have been already tested and used for previous Basel benchmarking exercises as well as for the QIS 1 of the Fundamental review of the trading book.
• Pre-validation exercise shall be straight forward within the short timeframe whereas it would be much more time consuming to book a whole set of new operations (option 1) and validate them.
• Starting from 2015 exercise, we could then switch to the new proposed portfolios.
Besides, we fully support the EBA’s proposition to organize a pre-validation phase to ensure correct booking of instruments. Before the start of any exercise, competent authorities should ensure that initial market valuations are within an acceptable range and questions concerning instructions are resolved. Previous exercises such as FRTB QIS on HPE showed that this pre-validation phase is of a key importance to limit operational risk.
In addition, we want to draw the EBA’s attention on the necessity for banks to receive portfolios specifications well in advance of the exercises. Booking positions in test portfolios, checking them and performing validation processes requires time which cannot be reduced. Finally, as far as possible, multiple iterations of instructions should be avoided to limit confusion or late changes to the portfolios.
Credit Risk
We generally favor the suggested approach to alternate exercises on LDP and HDP. A thorough examination of the results, followed by explanations between supervisors and banks will obviously be a lengthy process, and (at least on the credit risk side) loss trends emerge progressively: macroeconomic changes are the main risk drivers, and will take time to materialize in loss parameters.
We would like to focus our comments on the use of HPE for LDP: This approach seems inappropriate when it comes to estimate secured LGDs within the HPE exercise, for at least three main reasons:
• This hypothetical exercise would be a true hypothetical exercise (vs. previous HPE exercise that relies on existing information), since underlying transactions would not exist. Therefore, the reliability of estimates would be questionable and the workload would be much higher.
• There is very little chance that it is possible to provide all the requested information to determine CCF/LGD Secured. LGD secured is not a function of the collateral only: it depends on many different factors, including, obviously, the collateral (type, location, condition…) but also the ability to have access to it (seniority of the claim, legal environment, nature of the counterparty…), the characteristic of the loan (type, LTV…), the use of the asset…
• The outputs would depend on the zone of expertise of some institutions: for instance, a German bank would be able to give a LGD estimate for a German Real Estate transaction, whereas it would probably be much more difficult for a Retail Spanish bank.
The value of the approach is felt as limited, mainly owing to representativeness issues. We also would like to draw attention on the fact that cumulating this approach to the others for benchmarking purposes would become really burdensome for the banks on the LDP portfolios.
As for HDP, we consider that the portfolios defined for Mortgages are far too granular:
- While waiting that banks set up / adapt an adequate IT solution applied to the reporting for the benchmarking exercise, all of the requested information are not directly found in the current IT systems (ILTV buckets for instance might be recorded in another business or management IT environment);
- By selecting too many and too granular clusters, especially on retail exposures, the comparability purposes might turn out be difficult or vain, all the more so as these portfolios are more sensitive to local specificities;
- Some parties have also expressed concerns that these tiny clusters would not fit the range of application of the banks’ internal models: for instance one cluster could be covered by 2 models which makes the back-testing approach difficult to implement and interpret.
Is specialized lending included in the LDP sample and in particular for 2014? And if it is the case, which definition of specialized lending do we refer to (exposures with slotting criteria approach, type of portfolio, …)
Annex I
- Entity and registration identifiers (C101 – columns #390, 400, 410, 420, 430): please indicate how to fill in the available identifiers only (and not all of them) as banks may have different registration modes.
- Size of the counterparty (C102 - column #510): Please precise whether client turnover is on a solo or consolidated basis.
- For SME and Corporate portfolios, what does “Construction” mean in the ‘Portfolio name’ column (C103 – column #460): does it refer to the industry sector of the counterparty?
- For the residential mortgages portfolio (C103 - column #460), how do you intend to take into account the local specificities of types of real estate lending (for ex. Specify in the definition that it includes Credit Logement for France, Mandates for Belgium or NHG for Netherland, etc. and indicate which portfolio name it is split into).
Annex IV
- Column #220: “Default rate”: it is stated to see column “200 for the definition of default rate but column #200 refers to the “RWA standardized” (instead of #210?).
- It could be useful to set examples of calculation of RWA* and RWA** for both columns #250 and #260.
Annex V
Columns #640, 650 and 660: in many cases LGD models are not based on a multi-step modelling (i.e. distinguishing between cured and foreclosure), which makes it impossible to fill out the fields #640 (‘cure rate defaulted assets’), #650 (recovery rate not cured foreclosed assets’) nor #660 (‘recovery period length not cured foreclosed assets). At minimum, banks should be able to provide directly the LGD when they rely on a one-step approach as an alternative to the current set of required information.
The only portfolio we may think about for a second stage could cover:
- Plain vanilla debt securities held in the banking book;
- Equity holdings (since the latter are likely to be covered by BCBS Pillar 3 requirements).
- Booking: For all instruments, we highly recommend to use settlement dates that fall prior to the IMV reporting date in order to avoid comparison issues with other banks. Positions shall be booked at a specific date ahead of the pre-validation exercise and no update should be required on the IMV reporting date in order to minimize operational risks.
- IMV reporting time: Point 1b mentions timing at 4.30pm London time. However, for most participating banks, the valuation timing in systems depends on the product. We therefore recommend the EBA to leave banks flexibility in the timing of valuation and require them to report the valuation timing they used for each and single instrument.
Annex VII.a: EBA market risk benchmark portfolios
- Interest Rate: IRS swaps for GBP, EUR, SEK, DKK and USD are not standard (variable legs indexed on 1Y Ibor rates). As a result swaptions written on them are bespoke. What is the rationale for choosing non-standard swaps?
- Credit:
o Point 1i mentions IMM dates should be used. Does it mean IMM relative to the date of the exercise, or relative to 21 Feb 2014 ?
o The description of CDS is not always homogeneous, sometimes they include the seniority, sometimes they don’t (although the RED code is always added, hence seniority is implicit). In order to avoid any doubt, we recommend being systematic by providing always the seniority and RED Code. From September onwards there will be new ISDA definitions for CDSs. This might lead to confusion as to which products to book.
o Finally, for instrument #56 (AXA bond), the correct ISIN code is FR0011322668 and not FR001132266.
- Commodities:
o For instrument 32 (crude oil puts), could the EBA give the exact reference of the month on which the put is written (example: WTI Dec-15 future which has a last trading date in Nov-15) and the month used for the strike determination (example: Jun-15 WTI future which has a last trading date in May-15)?
o For instruments 33 and 34, we seek a few clarifications: Are they listed or OTC instruments? Are they ATM? Could the EBA provide specific dates to avoid any confusion?
- FX: To avoid any differences between banks, we recommend the EBA to define trades details directly rather than by reference:
o Dates: it would be easier to have the exact maturity, rather than 3M.
o Strike: Exact strikes could be defined directly rather by reference to “the rate published by the ECB on 28 February 2014” or “the price corresponding to the three-month forward exchange rate as of end of day 21 February 2014”.
We note some wording mistakes for instruments 30 and 31:
o 30. 3-month short forward DKK/USD currency, (short long DKK, long short USD EUR) with 1 USD Million purchased at the DKK/USD reference rate published by the ECB on 28 February 2014.
o 31. 3-month short forward SEK/USD currency, (long USD, short EUR SEK) with 1 USD Million purchased at the SEK/USD reference rate published by the ECB on 28 February 2014.
Sell call EUR put USD with strike = Current FX Forward x (1 + 1%) and sell put EUR call USD with strike = Current FX Forward x (1 - 1%).
In addition of the defined rules, we may suggest the following process in this perspective:
- The banks may select portfolios some of which it would be legitimate to exclude, based on a series of objective factors (to be defined by the supervisor: quantitative factors– see below, as well as qualitative – for instance a portfolio in run-off). A discussion would then take place between the supervisor and the bank to reach a conclusion on the opportunity to exclude the concerned portfolio(s), or maintain them in the analysis.
- Considering exemptions are granted, a special care should be taken regarding the aggregated portfolios analysis since exemptions could lead to consistency and comparability issues. Depending on the relative weight of the individual portfolios, when a bank is unable to provide results on a significant individual portfolio, we suggest it should be excluded from the peer’s distribution for aggregated portfolios that are impacted.
Such as:
- An absolute portfolio size ;
- A relative portfolio size, in comparison to the total consolidated balance-sheet or to the balance-sheet size of the subsidiary.
Portfolios with partial roll-out should also be exempted (or alternatively, the share of the portfolio under Standard approach should be highlighted and the bias resulting from different capital treatments should be eliminated).
Local entities supervised by a host supervisor should also be exempted from solo reporting as long as their portfolios are included in the consolidated vision submitted to the home supervisor.
Q2. Do you consider that the benchmarks outlined in the RTS are sufficiently proportionate and flexible? Do you have any alternative benchmark proposals? If yes, please provide details.
Irrespective of the appropriateness of the proposed benchmarks (discussed in answer to Q3), the question of proportion / flexibility is closely related to the obligations imposed on competent authorities when a contributor is deemed “outlier” with respect to the defined benchmarks.Would competent authorities be expected to investigate any “outlier” contribution and provide the EBA with a feedback, then we advocate the proposed benchmarks are overly prescriptive and disproportionate. For the sake of illustration, the “1st and 4th quartile” rule would systematically require competent authorities to enquire on half of the contributions. Similarly, the “80% SA output criteria” is likely to result in most contributions being considered as outlier for market risk test portfolios with diversification/netting features.
Alternative proposals are discussed in answers to questions 3 and 4.
Q3. What limitations do you see in relation to the use of the proposed benchmarks, i.e., (i) first and the fourth quartiles; (ii) comparison between own funds under the internal models and the standardised approach; and (iii) comparison between estimates and outturns?
First and fourth quartilesWe recognize the importance of using common and objective benchmarks to identify extreme values. However we consider the current criteria (1st and 4th quartile irrespective of the dispersion of the contributions) is unsuitable.
Starting from the EBA’s proposal an alternative approach could be:
Firstly in order not to get too many outlier banks, to consider other brackets quantiles: e.g. 1st and 9th deciles 0% and 90% instead of the 1st and 4th quartiles.
Secondly, among those outliers banks, to filter only those which absolute differences to the consensus (mean or rather median) are more important than a given threshold, which could be: a Euro amount, a percentage of a consensus on the portfolios mark-to-markets, a number of standard deviations (of the results distribution), etc.
We believe that selecting 50% of the sample (i.e. 1st and 4th quartiles) will create additional works for supervisors, EBA and banks with limited added value. We think outliers should be defined as banks ticking several criteria on the same portfolio, instead of the reunion of banks ticking at least one of the criteria.
At a minimum, the number of outlier contributions should depend on an indicator of dispersion. We recommend using the following formula:
n=max{F ,N ∙(α∙σ/Avg+β)}
Where:
n is such that the n smallest and n highest values are deemed outliers
N is the total number of contributions
σ is the standard deviation of the N contributions
Avg is the average of the N contributions
α and β are scalars to be calibrated
F is a floor level to be calibrated
Note that the above formula can be refined to account for asymmetry in the distance to average of above-average contributions and the distance to average of below-average contributions.
In addition, in case of bimodal distributions (which may happen for instance on some LGD models), the abrupt consideration of quartiles may prove inefficient. For instance, if a portfolio may result either in a high or low LGD with equal probability, Q1 and Q2 would correspond to the same value (high LGD), and so would Q3 and Q4 (low LGD).
We also fear that an inappropriate use of quartiles may result in the herd behavior the EBA wants to avoid: we can indeed assume there will be a strong pressure on banks to exit Q1 (assuming Q1 is the least conservative). On the other side, a position in Q4 is unlikely to result in decreases in overly conservative risk parameters (would banks positioned in Q4 be invited to renegotiate their margin of prudence or VaR multiplier…?).
Comparison between own funds under the internal models and the standardized approach
As developed in answer to question Q1, we reiterate our view that defining a common reference level of own funds at test portfolio level is inappropriate.
The appropriate level of capital for a given portfolio is firm-specific and must be assessed in the light of the firm’s ability to risk-manage such a portfolio, the internal model performance and the representativeness of the portfolio with respect to the overall internal model.
Regarding SA metrics, on top of being rough and non risk-sensitive there is not always unicity in its implementation: for instance on the credit risk side, see the diversity of national supervisory adjustments, or the treatment of sovereign risk under the permanent partial exemption in some countries. On market risk, within the course of SIGTB market risk benchmarking exercise Phase 2, CRM floor contributions were even more volatile than CRM contributions themselves. As a result, it is expected that SA contributions will be highly variable.
As far as market risk is concerned, should SA metrics be ultimately maintained as a common benchmark, we urge the EBA to consider at a minimum the following:
The application of SA output as a benchmark should be postponed until the FRTB new SA approach is implemented,
The computation of SA metrics should be performed at EBA level and the methodology used made available to all participating banks.
On the credit risk side, we are also unsure whether the EBA actually makes a direct reference to the standard approach, since our understanding of Article 4.1.a is that for the purpose of Art.78, sovereign exposures should be risk-weighted on the basis of the counterparty rating, thus disregarding potential existing local specific regulatory treatments - such as the exemption enabling banks to use a nil RW on a large sample of sovereign risks. Should this reading prove correct, we think this not only modifies the computation of Standard Approach as currently implemented by banks, but also opens the way to a new variety of alternative Standard computations, which may be required by various regulatory texts in the future: this would prove extremely complex, and would prevent from a direct comparison between the figures banks are used to reporting, and what they would then have to compute under such alternative Standard approaches. The additional operational risk arising from such manipulations has to be considered as well.
Comparison between estimates and outturns
EBA should give precise guidance on the computation of outturns (such as a spreadsheet), and these computations should take into account all components of the statistical validity of risk estimates, such as the number of years of available history, the default sample size, the number of models in use on the concerned cluster…
For example let us imagine a portfolio with decreasing origination over the recent years, because of careful risk management considerations: the cost of risk on such a portfolio is likely to be on the rise over a 5-year history, however the relative information brought by the 5 annual default rates is not identical due to the change in risk profile and risk appetite. How should such changes be reflected in the confidence interval?
Besides, according to us, the 97.5% level of confidence is a completely new paradigm which may prove inappropriate on some portfolios (owing to various sample sizes, or to the weight of expert judgment in the rating process) and extremely difficult and burdensome for banks to implement. We are not aware of any economic or academic studies proving the importance or relevance of such a confidence level and, therefore, we question the reasons why the EBA is putting forward this new standard. We strongly recommend that the calculation of RWA* and RWA** be abandoned due the many technical difficulties for banks to implement it and for competent authorities to interpret it.
Q4. What in your view is the most appropriate benchmark and/or approach for the assessment of the level of potential underestimation of own funds requirements?
Most valuable and meaningful approach should consist in identifying potential own funds requirements under-estimation, which is the one based on the comparison between model-estimated values and realized values, i.e. back-testing. Then again, competent authorities should use it in conjunction with other sources of investigation like the modelling construction and analysis as well as the validation history (local supervisory practices and decision on a particular model).Market risk
Our proposal consists in submitting market risk test portfolios to the back-testing test currently applied at trading book level. This is consistent with the Basel FRTB approach which introduces back testing at a much more granular level.
The flipside of the coin is that back-testing implies that the observation period is much longer than currently proposed. One possible option would be to make the exercise continuous over time while setting up clear rules for rolling tests periodically on underlying instruments of selected portfolios.
Credit risk
We understand the assessment of internal models is one of the objectives of the “General provisions” laid out in art. 8 of the RTS regarding the assessment of the credit risk internal models. Nevertheless we do not figure out accurately to what extent the competent authorities will make use of the information quoted in art. 8-1 and 8-2.
Besides, to avoid creating false outliers, Expected losses (%) or RW (%) should form the first level of comparison. Otherwise, two banks, one with a very early default definition and the other with a very late one, will tend to have high PD/low LGD and (respectively) low PD/high LGD whereas the results in EL% or RW% might look very similar.
Q5. Which set of market risk portfolios do you consider more appropriate for the initial exercise conducted under Article 78?
We are in favor of option 2 for the 2014 exercise.For the initial 2014 initial exercise, we highly recommend the use of the BCBS test portfolios. Indeed, we learnt from previous market risk benchmarking exercises that proper portfolio specification, consistent booking across participants and pre-validation exercise are key steps to ensure the exercise meets its target.
Given:
• Other ongoing regulatory exercises with colliding timelines (FRTB QIS, BCBS/EBA joint counterparty risk and CVA benchmarking exercise to mention only 2),
• The very tight timeframe and close kick-off date contemplated by the EBA for the 2014 exercise,
We consider conditions are not met to ensure the 2014 exercise can be properly performed on the basis of a new set of portfolios.
Q6. As explained in the background section, do you consider the approach proposed by the EBA appropriate for future annual exercises?
Market RiskAs described earlier, we favor the use of Basel portfolios for the first exercise to take place in Q4 2014 given that:
• The portfolios have been already tested and used for previous Basel benchmarking exercises as well as for the QIS 1 of the Fundamental review of the trading book.
• Pre-validation exercise shall be straight forward within the short timeframe whereas it would be much more time consuming to book a whole set of new operations (option 1) and validate them.
• Starting from 2015 exercise, we could then switch to the new proposed portfolios.
Besides, we fully support the EBA’s proposition to organize a pre-validation phase to ensure correct booking of instruments. Before the start of any exercise, competent authorities should ensure that initial market valuations are within an acceptable range and questions concerning instructions are resolved. Previous exercises such as FRTB QIS on HPE showed that this pre-validation phase is of a key importance to limit operational risk.
In addition, we want to draw the EBA’s attention on the necessity for banks to receive portfolios specifications well in advance of the exercises. Booking positions in test portfolios, checking them and performing validation processes requires time which cannot be reduced. Finally, as far as possible, multiple iterations of instructions should be avoided to limit confusion or late changes to the portfolios.
Credit Risk
We generally favor the suggested approach to alternate exercises on LDP and HDP. A thorough examination of the results, followed by explanations between supervisors and banks will obviously be a lengthy process, and (at least on the credit risk side) loss trends emerge progressively: macroeconomic changes are the main risk drivers, and will take time to materialize in loss parameters.
We would like to focus our comments on the use of HPE for LDP: This approach seems inappropriate when it comes to estimate secured LGDs within the HPE exercise, for at least three main reasons:
• This hypothetical exercise would be a true hypothetical exercise (vs. previous HPE exercise that relies on existing information), since underlying transactions would not exist. Therefore, the reliability of estimates would be questionable and the workload would be much higher.
• There is very little chance that it is possible to provide all the requested information to determine CCF/LGD Secured. LGD secured is not a function of the collateral only: it depends on many different factors, including, obviously, the collateral (type, location, condition…) but also the ability to have access to it (seniority of the claim, legal environment, nature of the counterparty…), the characteristic of the loan (type, LTV…), the use of the asset…
• The outputs would depend on the zone of expertise of some institutions: for instance, a German bank would be able to give a LGD estimate for a German Real Estate transaction, whereas it would probably be much more difficult for a Retail Spanish bank.
The value of the approach is felt as limited, mainly owing to representativeness issues. We also would like to draw attention on the fact that cumulating this approach to the others for benchmarking purposes would become really burdensome for the banks on the LDP portfolios.
As for HDP, we consider that the portfolios defined for Mortgages are far too granular:
- While waiting that banks set up / adapt an adequate IT solution applied to the reporting for the benchmarking exercise, all of the requested information are not directly found in the current IT systems (ILTV buckets for instance might be recorded in another business or management IT environment);
- By selecting too many and too granular clusters, especially on retail exposures, the comparability purposes might turn out be difficult or vain, all the more so as these portfolios are more sensitive to local specificities;
- Some parties have also expressed concerns that these tiny clusters would not fit the range of application of the banks’ internal models: for instance one cluster could be covered by 2 models which makes the back-testing approach difficult to implement and interpret.
Q7. Do you have any alternative proposals? If yes, please provide details.
For credit risk, an alternative proposal could be the use of country-specific industry means or medians together with a definition of a relative acceptable variation from the mean or median (to be defined by EBA or national competent authorities). This approach appears particularly relevant for the benchmarking exercises on mortgage portfolios. The divergences of RWA on this portfolio are mainly explained by the legal specificities (i.e. state guarantee and supervisory discretions).Q8. Which of the two options for phasing-in do you consider preferable?
We are in favor of option 2, provided that there is no significant evolution on portfolios except what could improve the relevance of these portfolios.Q9. Do you see any potential ambiguities in the credit risk portfolios defined in Annex I? Please identify the relevant portfolio providing details and any suggestions that would eliminate these ambiguities.
Consultation paper - Background and rationale – Credit risk (IRBA) p.10:Is specialized lending included in the LDP sample and in particular for 2014? And if it is the case, which definition of specialized lending do we refer to (exposures with slotting criteria approach, type of portfolio, …)
Annex I
- Entity and registration identifiers (C101 – columns #390, 400, 410, 420, 430): please indicate how to fill in the available identifiers only (and not all of them) as banks may have different registration modes.
- Size of the counterparty (C102 - column #510): Please precise whether client turnover is on a solo or consolidated basis.
- For SME and Corporate portfolios, what does “Construction” mean in the ‘Portfolio name’ column (C103 – column #460): does it refer to the industry sector of the counterparty?
- For the residential mortgages portfolio (C103 - column #460), how do you intend to take into account the local specificities of types of real estate lending (for ex. Specify in the definition that it includes Credit Logement for France, Mandates for Belgium or NHG for Netherland, etc. and indicate which portfolio name it is split into).
Annex IV
- Column #220: “Default rate”: it is stated to see column “200 for the definition of default rate but column #200 refers to the “RWA standardized” (instead of #210?).
- It could be useful to set examples of calculation of RWA* and RWA** for both columns #250 and #260.
Annex V
Columns #640, 650 and 660: in many cases LGD models are not based on a multi-step modelling (i.e. distinguishing between cured and foreclosure), which makes it impossible to fill out the fields #640 (‘cure rate defaulted assets’), #650 (recovery rate not cured foreclosed assets’) nor #660 (‘recovery period length not cured foreclosed assets). At minimum, banks should be able to provide directly the LGD when they rely on a one-step approach as an alternative to the current set of required information.
Q10. Do you have any suggestions for additional credit risk portfolios? Please provide details.
We support the EBA view to focus on plain vanilla credit instruments, since comparability is likely to be difficult to achieve on more sophisticated exposures, or exposures held by a limited number of contributing banks (such as securitization, private equity holdings…).The only portfolio we may think about for a second stage could cover:
- Plain vanilla debt securities held in the banking book;
- Equity holdings (since the latter are likely to be covered by BCBS Pillar 3 requirements).
Q11. Do you see any potential ambiguities in the market risk portfolios defined in Annexes VII.a and VII.b? Please identify the relevant portfolio providing details and any suggestions that would eliminate these.
Common Instructions- Booking: For all instruments, we highly recommend to use settlement dates that fall prior to the IMV reporting date in order to avoid comparison issues with other banks. Positions shall be booked at a specific date ahead of the pre-validation exercise and no update should be required on the IMV reporting date in order to minimize operational risks.
- IMV reporting time: Point 1b mentions timing at 4.30pm London time. However, for most participating banks, the valuation timing in systems depends on the product. We therefore recommend the EBA to leave banks flexibility in the timing of valuation and require them to report the valuation timing they used for each and single instrument.
Annex VII.a: EBA market risk benchmark portfolios
- Interest Rate: IRS swaps for GBP, EUR, SEK, DKK and USD are not standard (variable legs indexed on 1Y Ibor rates). As a result swaptions written on them are bespoke. What is the rationale for choosing non-standard swaps?
- Credit:
o Point 1i mentions IMM dates should be used. Does it mean IMM relative to the date of the exercise, or relative to 21 Feb 2014 ?
o The description of CDS is not always homogeneous, sometimes they include the seniority, sometimes they don’t (although the RED code is always added, hence seniority is implicit). In order to avoid any doubt, we recommend being systematic by providing always the seniority and RED Code. From September onwards there will be new ISDA definitions for CDSs. This might lead to confusion as to which products to book.
o Finally, for instrument #56 (AXA bond), the correct ISIN code is FR0011322668 and not FR001132266.
- Commodities:
o For instrument 32 (crude oil puts), could the EBA give the exact reference of the month on which the put is written (example: WTI Dec-15 future which has a last trading date in Nov-15) and the month used for the strike determination (example: Jun-15 WTI future which has a last trading date in May-15)?
o For instruments 33 and 34, we seek a few clarifications: Are they listed or OTC instruments? Are they ATM? Could the EBA provide specific dates to avoid any confusion?
- FX: To avoid any differences between banks, we recommend the EBA to define trades details directly rather than by reference:
o Dates: it would be easier to have the exact maturity, rather than 3M.
o Strike: Exact strikes could be defined directly rather by reference to “the rate published by the ECB on 28 February 2014” or “the price corresponding to the three-month forward exchange rate as of end of day 21 February 2014”.
We note some wording mistakes for instruments 30 and 31:
o 30. 3-month short forward DKK/USD currency, (short long DKK, long short USD EUR) with 1 USD Million purchased at the DKK/USD reference rate published by the ECB on 28 February 2014.
o 31. 3-month short forward SEK/USD currency, (long USD, short EUR SEK) with 1 USD Million purchased at the SEK/USD reference rate published by the ECB on 28 February 2014.
Q12. Do you have any suggestions for additional market risk portfolios? Please provide details.
For FX test portfolios, we propose to introduce a FX vanilla option out of the money with strike far from the FX forward by 2.33 standard-deviations. The goal is to check if the VaR captures the convexity between the current FX Spot and the 99% FX Spot bump: does the methodology use a Taylor approach or a full Revaluation? For the sake of illustration, the related instrument could be:Sell call EUR put USD with strike = Current FX Forward x (1 + 1%) and sell put EUR call USD with strike = Current FX Forward x (1 - 1%).
Q13 Do you agree with the possibility of allowing firms to refrain from reporting portfolios if one of the conditions stated in Article 3 is met?
We are in favor of the approach to exempt firms from reporting certain portfolios.In addition of the defined rules, we may suggest the following process in this perspective:
- The banks may select portfolios some of which it would be legitimate to exclude, based on a series of objective factors (to be defined by the supervisor: quantitative factors– see below, as well as qualitative – for instance a portfolio in run-off). A discussion would then take place between the supervisor and the bank to reach a conclusion on the opportunity to exclude the concerned portfolio(s), or maintain them in the analysis.
- Considering exemptions are granted, a special care should be taken regarding the aggregated portfolios analysis since exemptions could lead to consistency and comparability issues. Depending on the relative weight of the individual portfolios, when a bank is unable to provide results on a significant individual portfolio, we suggest it should be excluded from the peer’s distribution for aggregated portfolios that are impacted.
Q14 Do you have any suggestion about additional exemptions from reporting? If yes, please provide details.
It would be welcome to envisage also an exemption for credit risk. As for market risk, the exemptions should be based on situations where the institutions are under a model validation process for the portfolios included in the samples or for non-material portfolios. As such, a materiality threshold could be defined.Such as:
- An absolute portfolio size ;
- A relative portfolio size, in comparison to the total consolidated balance-sheet or to the balance-sheet size of the subsidiary.
Portfolios with partial roll-out should also be exempted (or alternatively, the share of the portfolio under Standard approach should be highlighted and the bias resulting from different capital treatments should be eliminated).
Local entities supervised by a host supervisor should also be exempted from solo reporting as long as their portfolios are included in the consolidated vision submitted to the home supervisor.