Response to second Joint Consultation on draft RTS on risk-mitigation techniques for OTC-derivative contracts not cleared by a CCP
Go back
The definition of variation margin (VM) refers to “outstanding contracts” and does not mention the fact that the positions may be covered by a netting agreement and thus combined to a single net position. Although this follows from context and the fact that netting agreements are expressly addressed in other provisions, it may be considered to clarify this in the definition as well.
- T+1 requirement IM (Art. 1 (3) EIM)
In the case the proposed timing would be imposed for the complete collection of margin, we think that this would be operationally unfeasible. Even for cash, market practice is as such that calculation could be performed in T and the call could be made in T+1 which would mean that the collateral would generally move not before T+2.
Consequently, the EBF reiterates its request to ensure the consistency of the timing imposed for calling and collecting initial and variation margins with the standard settlement/delivery regimes applicable to eligible collateral assets (i.e. broadly between one and three business days) with daily calculation and margin calls. This timeline enables a sound management of operational risk. This slight extension of the time-line would greatly reduce operational burden and associated operational risks.
An extended settlement period of T+3 would also help to address to some extent the considerable new complexities which will be introduced by the future RTS and which will further prolong the settlement process, such as the calculation of the MTA on group level as well as concentration limits. In addition the location of the parties and time zones in which they operate will need to be taken in consideration in this context: a cut-off time should be allowed to reflect the time zone differences between the parties´ respective locations.
At the very least it should be clarified that any breaches of the T+1 requirement where these are not caused by the party in question (that is where the causes are outside of its sphere, e.g. caused by the central securities depository) are not automatically deemed to constitute a breach of regulatory requirements.
- T+1 requirement VM (Art. 1 VM)
The above concerns apply correspondingly to the equally strict time limit (T+1) for exchanges of VM where no IM is being exchanged. As mentioned above, the time restrictions will be extremely challenging even in the case of cash collateral. They are impossible to observe in case of securities. The time limits for an exchange of VM should neither result in operationally challenging requirements nor should they effectively reduces usage of non-cash collateral for variation margin (mostly for the smaller entities solely exchanging variation margin)
Regarding the wording used in relation to the collection of variation margins, we would like to point out that the phrase settling exposures in cash" in Art. 1 VM (2)(a) should be replaced with the phrase "exchanging cash in amounts sufficient to extinguish exposures". The current language is not a correct general description of variation margins transfers for OTC derivatives because the transfer of cash variation margins does not necessarily settle current exposure. This change should also be reflected in Recital 11 (p.19).
- Settlement timeline
The distinction between the IM and VM settlement timelines disadvantages the smaller entities posting only variation margin and creates an uneven playing field with the bigger market participants. The distinction is not sensible from the operational perspective, since the smaller entities do not have the same infrastructure and liquidity as their bigger peers while facing shorter settlement deadlines. Also, the requirement discriminates the asset classes which may not be settled within 1 business day (such as Japanese government bonds or Yen cash).
It should also be recognised that certain entities may be restricted in what collateral type they can accept (by regulations, local market practices, internal policies etc.). Meeting the T+1 settlement deadline can then be challenging for their counterparties without much alternative collateral types to choose from. The liquidity issue has been taken into account when it comes to the concentration limits applicable to the governments bonds (where small non-systemically important institution where exempted). A consistent approach would be welcome here too.
- Application of the MTA when calculating VM and IM – Art. 4 (2) and (3) GEN
Recital 10 addresses the possibility that counterparties split the minimum transfer amount (MTA) in two separate amounts for VM and IM, respectively (the total of which not exceeding the prescribed maximum of 500.000 €). The provisions setting out the manner in which the amounts due are to be calculated when applying the MTA do, however, can be read to preclude a direct application of such a split MTA on the level of the VM and IM amounts since they appear to foresee an application of the MTA only after calculation of the IM and VM amounts., calculated in accordance with Art. 4 (2) (a) and (b) GEN. It should therefore be clarified that that in the case of a split MTA, these can be applied already directly on IM and VM level, that is when calculating the IM and VM amounts in accordance with Art. 4 (2) (a) and (b) GEN.
The authorised MTA may be subject to the currency fluctuations if the Base Currency is other than EUR. This seems to be a challenge for collateral management systems. The currency effects should be allowed to the extent the MTA was in line with the maximum authorised cap at the time the agreement was entered into between the parties.
- Application of group IM threshold – Art. 6 GEN
The provisions regarding the application of the threshold on group level do not clearly set out that this threshold is to be applied in relation to the own group and not the group of the counterparty, that is, each group is responsible for monitoring its overall exposure to a specific counterparty and have procedures in place in order to ascertain that the own group members observe the group threshold when entering into transactions with this specific counterparty. This could be clarified in a recital to avoid confusion."
Regarding the Article 3 MRM, the EBF would welcome clarity on the definition of “a period of significant financial stress” and how this will be interpreted in practice.
The EBF welcomes the confirmation by the ESAs that counterparties will be entitled to use internal initial margin models. The EBF also welcomes the greater flexibility granted by the ESAs on some issues, and notably the consent by the ESAs to a follow-up validation of initial margin models on a yearly basis (unless significant changes within this time-period) and not twice a year.
It should be possible to choose between either an approach based on assigning a derivative contract to an underlying risk class based on its primary risk factor or an approach which entails the calculation of all risk factors (interest rate, equity, etc.) for all trades and arriving at aggregate numbers for these risk factors. The latter approach would, for example, ensure that the interest rate risk is still calculated and included for trades where, for example, equity maybe the primary risk factor.
The RTS states that initial margin models should be based on an asset-class comparison (i.e. a comparison with OTC derivatives that are in the same “netting set” and within the same “underlying asset class” as defined in the draft RTS). However some banks consider that initial margin models should be determined on a risk-based approach (i.e. via a classification by type and degree of risk), as the latter criterion appears to be more consistent with the calculation of appropriate initial margins.
The assessment of the liquidity of the derivative contracts portfolio (required in order to assess the validity of initial margin models) may raise various issues: (i) the estimation of such liquidity is very problematic as the liquidity may evolve and does not rely on standardised predetermined criteria, (ii) the assessment of the liquidity will imply operational constraints for banks as the latter will define sub-categories of derivative contracts portfolios and calculate separate and specific initial margins for transactions concerning each of these sub-categories, and (iii) this sub-categorisation of portfolios may lead to an increase of the global initial margins to be posted/collected (securing each sub-category of derivative contracts portfolios may be more costly for each counterparty than securing a global derivative contracts portfolio).
Concentration limits on eligible collateral is not a feature of the BCBS-IOSCO framework document. This may dissuade non-EU entities from trading uncleared OTC derivatives with EU entities. Concentration limits should be defined in such a way that it only defines limits to ensure that the value of and ability to liquidate the collateral is secured in the event of a counterparty default. With this in mind we do not see that restricting equities to 40% of posted IM is reflective of the liquidity characteristics of equities, especially since the criteria further restricts eligible equities to only those from the main indices. We would note that even during the financial crisis good levels of liquidity was maintained in the equity markets. We would therefore suggest that the concentration limit (40%) should be significantly higher and possibly removed.
In addition, the treatment of covered/mortgage bonds seems unnecessarily restrictive and in our view does not reflect the treatment of covered bonds under other EU regulation where they are considered high quality liquid asset, practically in-line with government bonds. We would refer specifically to studies performed by the EBA in the context of the eligibility of such asset within banks liquidity buffers, for LCR.
Another point of note is that for some jurisdictions there is a potential shortfall of eligible government bonds. In some Member States the size of the in covered bond market is greater than the size of the government bond market, elevating the significance of and therefore the need for similar treatment of covered bonds (under EMIR), given their (observed) similar quality and liquidity characterises to government bonds.
One potential consequence of an overly punitive treatment (e.g. the maximum contribution from any one issuer set to 10% of the total collateral value), combined with a shortage of other equivalent assets, would be that other lower quality and less liquid assets could be posted as collateral, i.e. an overall collateral “downgrade”, increasing risks for counterparties. We would therefore propose that under EMIR that covered/mortgage bonds receive similar treatment as government bonds, i.e. to have no concentration limits or that the maximum concentration limit is raised considerably.
We welcome the exemption for government and municipal bonds but this requires yet another monitoring procedure of own/counterparty´s SSI status and IM levels. The principle of proportionality (i.e. to avoid too cumbersome obligation given the limited regulatory benefit) seems to require that the criteria for the exemption is based on an existing categorisation or data already imposed to be exchanged otherwise (such as IM trigger). If this monitoring would be too cumbersome, parties will be discouraged from posting high quality collateral.
For consistency with the application of the concentration limits, the threshold in paragraph 3(c) shall be applicable at the individual counterparty entity level and not at group basis.
We note that the mechanism through which the SSI status will be monitored and published has not yet been implemented by the authorities. It may therefore be potentially difficult to monitor at the moment.
We would like to propose raising the EUR 1 billion of collected collateral threshold. For smaller institutions, a handful of relationship which may reach this level would drag then into the cumbersome monitoring procedures and systems solutions. Alternatively, we would suggest a general exemption from the proposed concentration limits for counterparties with low exposure, posting less than EUR 100 million.
The 1 billion euro threshold raises some specific issues: The calculation modalities of the 1 billion euro threshold should be clarified: shall this threshold be calculated individually (i.e. by netting set as specified during the EBA’s public hearing) or globally (i.e. taking into account all the collateral exchanged with all the counterparties of the bank)? The second option seems more consistent with the purpose of the regulation – and would be more practicable from an operational viewpoint.
The calculation period of the 1 billion euro threshold should also be clarified: shall it be calculated on a daily basis or over a specified time-period? The second option would be more practicable from an operational viewpoint.
The monitoring of concentration limits with respect to the IM segregated at a third party custodian presumes that the custodians will assist with the monitoring.
Although the ESAs noted the intention to align with the international standards, the requirement to monitor the concentration limits and wrong way risk of the collected collateral remains at odds with the US margining regime.
The ESAs should also clarify the consequences of a potential evolution of the amount of the collateral exchanged with counterparties just above or just under the said 1 billion euro threshold, notably in cases where this evolution takes place within the timeframe of the threshold calculation period.
Regarding the need for entities which are not qualified as G-SIIs and may not diversify their collateral offered, the EBF also recalls that such proposal will require from the banks to categorise their clients into tiers (a tier 1 covering clients benefiting of the exemption mentioned, and a tier 2 covering the non-exempted clients), which would impose a significant operational burden for the banks.
The requirement for the legal analysis of the netting arrangements is new, introduced in the second consultation alongside the original requirement for analysis of the validity of segregation arrangements. However, we believe that the requirement with respect to netting is already satisfactorily addressed through the capital adequacy framework as precondition for netting benefit under the Article 296 of Regulation (EU) No 575/2013. The RTS shall not introduce a supplementary legal review process which is not fully consistent with the capital adequacy framework process.
We understand the relevance of analysis of validity and enforceability of the netting arrangements. In some jurisdictions however, including the EU Member States, a clean netting opinion cannot be obtained. The status cannot be remedied unless the local bankruptcy legislation is amended to recognize the close-out netting arrangements upon opening of insolvency proceedings of the local counterparties.
We refer to our response on Q1 regarding the need to permit alternative approaches in the case a local jurisdiction does not support netting.
We further welcome the clarification made in the Public Hearing that the term “independent” is to be interpreted in accordance with the EBA response to Question No. 2013_23 in the Single Rulebook Q&A process; thus including provision of qualified internal legal advice, i.e. from the legal department of the institution.
The changes and clarifications made provide for the necessary degree of flexibility regarding the manner in which a legal review process can be implemented as part of and under the existing risk management systems and procedures of a counterparty.
For example, we assume that the independent legal review requirement can been deemed to have been met, where a counterparty has procedures in place which provide for a monitoring / regular assessment whether there have been any material changes in the relevant laws which may have adversely affected the enforceability and validity of the contractual arrangements used, and where such material changes then trigger further steps which may include, for example, the obtaining
The requirement to conduct the legal review annually is still cumbersome and perhaps not feasible with current legal resources available to the participants in-house. The operational and cost implications are therefore significant. We would suggest aligning the frequency of the review to the position of the Article 296 of Regulation (EU) No 575/2013 leaving the assessment to the internal policies on “if and when required” basis.
EBF considers that such requirement could increase the systemic risk in the transactions with counterparties located in jurisdictions where the close-out netting is not enforceable and would recommend to remove the obligation to post collateral in the context of transactions entered into with counterparties domiciled in jurisdictions in which the close-out netting would not be enforceable upon insolvency of a local party or where it is not possible to guarantee the effectiveness of the segregation agreements.
In Article 2 OPD – Trading Documentation, page 47, clause 1.d. clarification and definition is needed for the term “calculation methods”.
- Written trading documentation (Art. 2 (1) OPD)
The term “trading documentation” is open to interpretation. We assume that this is intended to cover the contractual documentation setting out the general legal basis for the transactions to be concluded, namely the general framework for the margin requirements (including supplemental agreement) as well any netting agreement, but does not include the trade confirmation with the commercial details and sometimes further specific terms governing the specific transactions. This would reflect the established market practice of relying on standardised contractual agreements which address the general framework and key provisions (such as netting provisions and collateral annexes) and the use of trade confirmations to determine further details. The general framework (master agreement and collateral annex) is generally entered into in advance or on the conclusion of a transaction whereas the trade confirmations, by their nature, are issued subsequently following the agreement on the specific terms (which agreement can occur many forms, including by telephone). Consequently, a clear distinction has to be drawn between the general contractual framework which has to be in place in advance of or at the time of the conclusion of the transaction on the one hand and the trade confirmation which may cover additional terms but may only be formally documented following the agreement on a transaction on the other.
The use of the word “any” in Art. 2 (1) (a) OPD can be understood to mean that the relevant contractual documentation needs to include provisions which cover every possible payment obligation which may arise under or in relation to a transaction and/or contractual relationship. Such an understanding would be incorrect since contractual documentation intends or is able to specify or cover all possible payment obligations which emanate or may emanate from a contractual relationship. In addition, this understanding would conflict with the above described situation that certain transaction specific terms including payments may be agreed and documented in the trade confirmation, and thus not already in advance of a transaction. The contractual documentation will only be able to cover the general or material terms and the general basis for the rights and obligations but not any and all rights and obligations. The term “any” should therefore be deleted. The above applies correspondingly to the use of the term “any” in Art. 2 (1) (e).
The use of the term “written” in this context can be understood to preclude electronic messages and means of recording/documenting. It should therefore either be deleted or replaced by a term with a broader meaning covering any form which ensures an adequate recording/documentation, including electronic records.
The concerns already raised above in our response to Question 5 regarding the legal review requirement (repeated here for the sake of convenience) apply correspondingly. We would like to raise the same concerns with respect to the performance of the annual independent legal review as in Question 5 above.
- Segregation of initial margins (Art. 1 SEG)
The introduction of the new requirement for the IM to be protected from the default or insolvency of the “third party holder or custodian” amounts to in practice prohibit posting of cash as IM. Indeed, of the cash is inherently linked to the solvency of the third party bank/custodian account holder. In line with the Art. 1 REU, the cash will only be permissible as initial margin if it is re-invested. The consequential prohibition of initial margin in cash unless it is transformed into an alternative asset is too restrictive as cash was expressly envisaged by the BCBS-IOSCO framework for margin requirements for non-centrally cleared derivatives as a major liquidity safety net where securities are not available (distressed markets).
The Art. 1 SEG seems to describe only a model of third party segregation at the level of the collecting party (i.e. held segregated in the name of the collecting party). However, it does not expressly envisage segregation on an account at a third party banks in the name of the posting party that is charged in favour of the collecting party. Indeed, the second model does not involve segregation from proprietary assets of collecting party described and required in Art. 1 SEG (1) to (2). It seems even prohibited by the Art. 2 LEC (d) whereby the cash IM shall be maintained with a party other than collateral provider. This second model is however one of the segregation models in development which is expected to be used in most cases by the market participants.
In that context, in accordance with the BCBS-IOSCO recommendations, it should be made clear that in addition to the third-party holder or custodian initial margins could be collected by the collecting counterparty provided that the latter complies with the requirements of Articles 1 LEC and 1 SEG. We think that the first paragraph of Article 1 SEG does not provide enough clarity. In light of the above we suggest that “collateral collected as initial margin shall be segregated from proprietary assets on the books and records of the collecting counterparty, a third party holder or custodian, or via other legally binding arrangements made by the collecting counterparty to protect the initial margin from the default or insolvency of the collecting counterparty, third party holder or custodian.”
Finally we believe the requirement for re-investment of cash IM, if maintained, shall exempt the “incidental cash” such as distributions, redemptions and other cash generated by the re-investment).
- Relation between segregation requirements and collateral management requirements (Art. 1 SEG and Art. 2 LEC)
Art. 2 LEC appears to address some aspect which are also covered by the segregation requirements. One example is the requirement under Art. 2 (1) (c) LEC regarding insolvency or bankruptcy remoteness of initial margin maintained with the collateral provider. It is not entirely clear how these obligations covering similar or connected questions are to interact or which requirements prevail in the case of conflict. In particular all requirements addressing insolvency/bankruptcy remoteness need to be coordinated to avoid discrepancies or uncertainties. To this end, these should only be covered by one single provision.
- Assessment of the effectiveness of the netting and segregation agreements
As to the need to for alternative approaches and deviations from the obligation to exchange collateral in the case that segregation agreement and/or netting agreements are not sufficiently protected and supported by the laws of a third country, see our comments regarding this issue in our response to Q1 and Q5 above.
The reinvestment solution in Article 1 REU (2) may not be workable from a practical/commercial perspective. It presumes existing of a framework whereby the custodians would conduct the reinvestment according to the strategy agreed with the parties and moreover in compliance with the credit quality and concentration limits requirements does not currently exists. This therefore does not seem satisfactory to facilitate the usage of cash as IM to the extent it cannot be posted without being re-invested.
The proposal is also too restrictive for other reasons. Indeed, the cash re-used must be reinvested “only for the purpose of protecting the collateral poster” in case of default of the collateral beneficiary. For such purpose, ESAs specify that the cash used must be reinvested in securities and that the latter have to be segregated and not re-used. This restricted discretion granted to the collateral beneficiary may challenge the interest of such re-use while a huge portion of the collateral will be posted in cash.
The EBF welcomes that no FX haircut will apply where VM is posted in cash. However, the FX haircuts applicable in all other cases creates prohibitive cost or alternatively an important operational burden on the collateral management should the collateral be posted in currency silos. We support the previous request to replace 8% haircut entirely and include currency mismatches in the IM model for those parties that collect IM.
Regarding FX currency mismatch it is inconsistent that the RTS exempt physically settled FX derivatives from the IM requirements in Art. 5 GEN, but at the same time introduce an 8% haircut as mentioned in paragraph 6 and 7 on page 57. Haircut addresses potential future loss on the value of the collateral. Given that the majority of the risk in a transaction typically is constituted by the FX transaction, it seems unnecessary to introduce haircut on collateral. Furthermore, the haircut of 8% ought to be lower for currency pairs with low volatility, e.g. when one currency is pegged to another.
Respondents are invited to comment on the proposal in this section concerning the timing of calculation, call and delivery of initial and variation margins.
- Definition of variation margin (Art. 1 (5) GEN)The definition of variation margin (VM) refers to “outstanding contracts” and does not mention the fact that the positions may be covered by a netting agreement and thus combined to a single net position. Although this follows from context and the fact that netting agreements are expressly addressed in other provisions, it may be considered to clarify this in the definition as well.
- T+1 requirement IM (Art. 1 (3) EIM)
In the case the proposed timing would be imposed for the complete collection of margin, we think that this would be operationally unfeasible. Even for cash, market practice is as such that calculation could be performed in T and the call could be made in T+1 which would mean that the collateral would generally move not before T+2.
Consequently, the EBF reiterates its request to ensure the consistency of the timing imposed for calling and collecting initial and variation margins with the standard settlement/delivery regimes applicable to eligible collateral assets (i.e. broadly between one and three business days) with daily calculation and margin calls. This timeline enables a sound management of operational risk. This slight extension of the time-line would greatly reduce operational burden and associated operational risks.
An extended settlement period of T+3 would also help to address to some extent the considerable new complexities which will be introduced by the future RTS and which will further prolong the settlement process, such as the calculation of the MTA on group level as well as concentration limits. In addition the location of the parties and time zones in which they operate will need to be taken in consideration in this context: a cut-off time should be allowed to reflect the time zone differences between the parties´ respective locations.
At the very least it should be clarified that any breaches of the T+1 requirement where these are not caused by the party in question (that is where the causes are outside of its sphere, e.g. caused by the central securities depository) are not automatically deemed to constitute a breach of regulatory requirements.
- T+1 requirement VM (Art. 1 VM)
The above concerns apply correspondingly to the equally strict time limit (T+1) for exchanges of VM where no IM is being exchanged. As mentioned above, the time restrictions will be extremely challenging even in the case of cash collateral. They are impossible to observe in case of securities. The time limits for an exchange of VM should neither result in operationally challenging requirements nor should they effectively reduces usage of non-cash collateral for variation margin (mostly for the smaller entities solely exchanging variation margin)
Regarding the wording used in relation to the collection of variation margins, we would like to point out that the phrase settling exposures in cash" in Art. 1 VM (2)(a) should be replaced with the phrase "exchanging cash in amounts sufficient to extinguish exposures". The current language is not a correct general description of variation margins transfers for OTC derivatives because the transfer of cash variation margins does not necessarily settle current exposure. This change should also be reflected in Recital 11 (p.19).
- Settlement timeline
The distinction between the IM and VM settlement timelines disadvantages the smaller entities posting only variation margin and creates an uneven playing field with the bigger market participants. The distinction is not sensible from the operational perspective, since the smaller entities do not have the same infrastructure and liquidity as their bigger peers while facing shorter settlement deadlines. Also, the requirement discriminates the asset classes which may not be settled within 1 business day (such as Japanese government bonds or Yen cash).
It should also be recognised that certain entities may be restricted in what collateral type they can accept (by regulations, local market practices, internal policies etc.). Meeting the T+1 settlement deadline can then be challenging for their counterparties without much alternative collateral types to choose from. The liquidity issue has been taken into account when it comes to the concentration limits applicable to the governments bonds (where small non-systemically important institution where exempted). A consistent approach would be welcome here too.
- Application of the MTA when calculating VM and IM – Art. 4 (2) and (3) GEN
Recital 10 addresses the possibility that counterparties split the minimum transfer amount (MTA) in two separate amounts for VM and IM, respectively (the total of which not exceeding the prescribed maximum of 500.000 €). The provisions setting out the manner in which the amounts due are to be calculated when applying the MTA do, however, can be read to preclude a direct application of such a split MTA on the level of the VM and IM amounts since they appear to foresee an application of the MTA only after calculation of the IM and VM amounts., calculated in accordance with Art. 4 (2) (a) and (b) GEN. It should therefore be clarified that that in the case of a split MTA, these can be applied already directly on IM and VM level, that is when calculating the IM and VM amounts in accordance with Art. 4 (2) (a) and (b) GEN.
The authorised MTA may be subject to the currency fluctuations if the Base Currency is other than EUR. This seems to be a challenge for collateral management systems. The currency effects should be allowed to the extent the MTA was in line with the maximum authorised cap at the time the agreement was entered into between the parties.
- Application of group IM threshold – Art. 6 GEN
The provisions regarding the application of the threshold on group level do not clearly set out that this threshold is to be applied in relation to the own group and not the group of the counterparty, that is, each group is responsible for monitoring its overall exposure to a specific counterparty and have procedures in place in order to ascertain that the own group members observe the group threshold when entering into transactions with this specific counterparty. This could be clarified in a recital to avoid confusion."
Respondent are invited to provide comments on whether the draft RTS might produce unintended consequence concerning the design or the implementation of initial margin models.
- Calibration of the model (Art. 3 MRM)Regarding the Article 3 MRM, the EBF would welcome clarity on the definition of “a period of significant financial stress” and how this will be interpreted in practice.
The EBF welcomes the confirmation by the ESAs that counterparties will be entitled to use internal initial margin models. The EBF also welcomes the greater flexibility granted by the ESAs on some issues, and notably the consent by the ESAs to a follow-up validation of initial margin models on a yearly basis (unless significant changes within this time-period) and not twice a year.
It should be possible to choose between either an approach based on assigning a derivative contract to an underlying risk class based on its primary risk factor or an approach which entails the calculation of all risk factors (interest rate, equity, etc.) for all trades and arriving at aggregate numbers for these risk factors. The latter approach would, for example, ensure that the interest rate risk is still calculated and included for trades where, for example, equity maybe the primary risk factor.
The RTS states that initial margin models should be based on an asset-class comparison (i.e. a comparison with OTC derivatives that are in the same “netting set” and within the same “underlying asset class” as defined in the draft RTS). However some banks consider that initial margin models should be determined on a risk-based approach (i.e. via a classification by type and degree of risk), as the latter criterion appears to be more consistent with the calculation of appropriate initial margins.
The assessment of the liquidity of the derivative contracts portfolio (required in order to assess the validity of initial margin models) may raise various issues: (i) the estimation of such liquidity is very problematic as the liquidity may evolve and does not rely on standardised predetermined criteria, (ii) the assessment of the liquidity will imply operational constraints for banks as the latter will define sub-categories of derivative contracts portfolios and calculate separate and specific initial margins for transactions concerning each of these sub-categories, and (iii) this sub-categorisation of portfolios may lead to an increase of the global initial margins to be posted/collected (securing each sub-category of derivative contracts portfolios may be more costly for each counterparty than securing a global derivative contracts portfolio).
Respondents are invited to comment on whether the requirements of this section concerning the concentration limits address the concerns expressed on the previous proposal.
- Concentration limits (Art. 7 LEC)Concentration limits on eligible collateral is not a feature of the BCBS-IOSCO framework document. This may dissuade non-EU entities from trading uncleared OTC derivatives with EU entities. Concentration limits should be defined in such a way that it only defines limits to ensure that the value of and ability to liquidate the collateral is secured in the event of a counterparty default. With this in mind we do not see that restricting equities to 40% of posted IM is reflective of the liquidity characteristics of equities, especially since the criteria further restricts eligible equities to only those from the main indices. We would note that even during the financial crisis good levels of liquidity was maintained in the equity markets. We would therefore suggest that the concentration limit (40%) should be significantly higher and possibly removed.
In addition, the treatment of covered/mortgage bonds seems unnecessarily restrictive and in our view does not reflect the treatment of covered bonds under other EU regulation where they are considered high quality liquid asset, practically in-line with government bonds. We would refer specifically to studies performed by the EBA in the context of the eligibility of such asset within banks liquidity buffers, for LCR.
Another point of note is that for some jurisdictions there is a potential shortfall of eligible government bonds. In some Member States the size of the in covered bond market is greater than the size of the government bond market, elevating the significance of and therefore the need for similar treatment of covered bonds (under EMIR), given their (observed) similar quality and liquidity characterises to government bonds.
One potential consequence of an overly punitive treatment (e.g. the maximum contribution from any one issuer set to 10% of the total collateral value), combined with a shortage of other equivalent assets, would be that other lower quality and less liquid assets could be posted as collateral, i.e. an overall collateral “downgrade”, increasing risks for counterparties. We would therefore propose that under EMIR that covered/mortgage bonds receive similar treatment as government bonds, i.e. to have no concentration limits or that the maximum concentration limit is raised considerably.
We welcome the exemption for government and municipal bonds but this requires yet another monitoring procedure of own/counterparty´s SSI status and IM levels. The principle of proportionality (i.e. to avoid too cumbersome obligation given the limited regulatory benefit) seems to require that the criteria for the exemption is based on an existing categorisation or data already imposed to be exchanged otherwise (such as IM trigger). If this monitoring would be too cumbersome, parties will be discouraged from posting high quality collateral.
For consistency with the application of the concentration limits, the threshold in paragraph 3(c) shall be applicable at the individual counterparty entity level and not at group basis.
We note that the mechanism through which the SSI status will be monitored and published has not yet been implemented by the authorities. It may therefore be potentially difficult to monitor at the moment.
We would like to propose raising the EUR 1 billion of collected collateral threshold. For smaller institutions, a handful of relationship which may reach this level would drag then into the cumbersome monitoring procedures and systems solutions. Alternatively, we would suggest a general exemption from the proposed concentration limits for counterparties with low exposure, posting less than EUR 100 million.
The 1 billion euro threshold raises some specific issues: The calculation modalities of the 1 billion euro threshold should be clarified: shall this threshold be calculated individually (i.e. by netting set as specified during the EBA’s public hearing) or globally (i.e. taking into account all the collateral exchanged with all the counterparties of the bank)? The second option seems more consistent with the purpose of the regulation – and would be more practicable from an operational viewpoint.
The calculation period of the 1 billion euro threshold should also be clarified: shall it be calculated on a daily basis or over a specified time-period? The second option would be more practicable from an operational viewpoint.
The monitoring of concentration limits with respect to the IM segregated at a third party custodian presumes that the custodians will assist with the monitoring.
Although the ESAs noted the intention to align with the international standards, the requirement to monitor the concentration limits and wrong way risk of the collected collateral remains at odds with the US margining regime.
The ESAs should also clarify the consequences of a potential evolution of the amount of the collateral exchanged with counterparties just above or just under the said 1 billion euro threshold, notably in cases where this evolution takes place within the timeframe of the threshold calculation period.
Regarding the need for entities which are not qualified as G-SIIs and may not diversify their collateral offered, the EBF also recalls that such proposal will require from the banks to categorise their clients into tiers (a tier 1 covering clients benefiting of the exemption mentioned, and a tier 2 covering the non-exempted clients), which would impose a significant operational burden for the banks.
Respondent to this consultation are invited to highlight their concerns on the requirements on trading relationship documentation.
We welcome the ESAs proposal in not placing an undue burden on FCs and NFC+s in order to obtain a written agreement with all counterparties, including NFC-s. This will reduce the unnecessary documentation and operation burden on banks dealing uncleared OTC derivatives with corporates. FCs and NFC+s can then focus on their preparation and compliance efforts on transactions that are in scope of the BCBS-IOSCO framework i.e. uncleared OTC derivatives with and between FCs and NFC+s.The requirement for the legal analysis of the netting arrangements is new, introduced in the second consultation alongside the original requirement for analysis of the validity of segregation arrangements. However, we believe that the requirement with respect to netting is already satisfactorily addressed through the capital adequacy framework as precondition for netting benefit under the Article 296 of Regulation (EU) No 575/2013. The RTS shall not introduce a supplementary legal review process which is not fully consistent with the capital adequacy framework process.
We understand the relevance of analysis of validity and enforceability of the netting arrangements. In some jurisdictions however, including the EU Member States, a clean netting opinion cannot be obtained. The status cannot be remedied unless the local bankruptcy legislation is amended to recognize the close-out netting arrangements upon opening of insolvency proceedings of the local counterparties.
We refer to our response on Q1 regarding the need to permit alternative approaches in the case a local jurisdiction does not support netting.
We further welcome the clarification made in the Public Hearing that the term “independent” is to be interpreted in accordance with the EBA response to Question No. 2013_23 in the Single Rulebook Q&A process; thus including provision of qualified internal legal advice, i.e. from the legal department of the institution.
The changes and clarifications made provide for the necessary degree of flexibility regarding the manner in which a legal review process can be implemented as part of and under the existing risk management systems and procedures of a counterparty.
For example, we assume that the independent legal review requirement can been deemed to have been met, where a counterparty has procedures in place which provide for a monitoring / regular assessment whether there have been any material changes in the relevant laws which may have adversely affected the enforceability and validity of the contractual arrangements used, and where such material changes then trigger further steps which may include, for example, the obtaining
The requirement to conduct the legal review annually is still cumbersome and perhaps not feasible with current legal resources available to the participants in-house. The operational and cost implications are therefore significant. We would suggest aligning the frequency of the review to the position of the Article 296 of Regulation (EU) No 575/2013 leaving the assessment to the internal policies on “if and when required” basis.
EBF considers that such requirement could increase the systemic risk in the transactions with counterparties located in jurisdictions where the close-out netting is not enforceable and would recommend to remove the obligation to post collateral in the context of transactions entered into with counterparties domiciled in jurisdictions in which the close-out netting would not be enforceable upon insolvency of a local party or where it is not possible to guarantee the effectiveness of the segregation agreements.
In Article 2 OPD – Trading Documentation, page 47, clause 1.d. clarification and definition is needed for the term “calculation methods”.
- Written trading documentation (Art. 2 (1) OPD)
The term “trading documentation” is open to interpretation. We assume that this is intended to cover the contractual documentation setting out the general legal basis for the transactions to be concluded, namely the general framework for the margin requirements (including supplemental agreement) as well any netting agreement, but does not include the trade confirmation with the commercial details and sometimes further specific terms governing the specific transactions. This would reflect the established market practice of relying on standardised contractual agreements which address the general framework and key provisions (such as netting provisions and collateral annexes) and the use of trade confirmations to determine further details. The general framework (master agreement and collateral annex) is generally entered into in advance or on the conclusion of a transaction whereas the trade confirmations, by their nature, are issued subsequently following the agreement on the specific terms (which agreement can occur many forms, including by telephone). Consequently, a clear distinction has to be drawn between the general contractual framework which has to be in place in advance of or at the time of the conclusion of the transaction on the one hand and the trade confirmation which may cover additional terms but may only be formally documented following the agreement on a transaction on the other.
The use of the word “any” in Art. 2 (1) (a) OPD can be understood to mean that the relevant contractual documentation needs to include provisions which cover every possible payment obligation which may arise under or in relation to a transaction and/or contractual relationship. Such an understanding would be incorrect since contractual documentation intends or is able to specify or cover all possible payment obligations which emanate or may emanate from a contractual relationship. In addition, this understanding would conflict with the above described situation that certain transaction specific terms including payments may be agreed and documented in the trade confirmation, and thus not already in advance of a transaction. The contractual documentation will only be able to cover the general or material terms and the general basis for the rights and obligations but not any and all rights and obligations. The term “any” should therefore be deleted. The above applies correspondingly to the use of the term “any” in Art. 2 (1) (e).
The use of the term “written” in this context can be understood to preclude electronic messages and means of recording/documenting. It should therefore either be deleted or replaced by a term with a broader meaning covering any form which ensures an adequate recording/documentation, including electronic records.
Respondents are invited to comment on the requirements of this section concerning the legal basis for the compliance.
- Independent legal review requirement (Art. 1 (5) SEG)The concerns already raised above in our response to Question 5 regarding the legal review requirement (repeated here for the sake of convenience) apply correspondingly. We would like to raise the same concerns with respect to the performance of the annual independent legal review as in Question 5 above.
- Segregation of initial margins (Art. 1 SEG)
The introduction of the new requirement for the IM to be protected from the default or insolvency of the “third party holder or custodian” amounts to in practice prohibit posting of cash as IM. Indeed, of the cash is inherently linked to the solvency of the third party bank/custodian account holder. In line with the Art. 1 REU, the cash will only be permissible as initial margin if it is re-invested. The consequential prohibition of initial margin in cash unless it is transformed into an alternative asset is too restrictive as cash was expressly envisaged by the BCBS-IOSCO framework for margin requirements for non-centrally cleared derivatives as a major liquidity safety net where securities are not available (distressed markets).
The Art. 1 SEG seems to describe only a model of third party segregation at the level of the collecting party (i.e. held segregated in the name of the collecting party). However, it does not expressly envisage segregation on an account at a third party banks in the name of the posting party that is charged in favour of the collecting party. Indeed, the second model does not involve segregation from proprietary assets of collecting party described and required in Art. 1 SEG (1) to (2). It seems even prohibited by the Art. 2 LEC (d) whereby the cash IM shall be maintained with a party other than collateral provider. This second model is however one of the segregation models in development which is expected to be used in most cases by the market participants.
In that context, in accordance with the BCBS-IOSCO recommendations, it should be made clear that in addition to the third-party holder or custodian initial margins could be collected by the collecting counterparty provided that the latter complies with the requirements of Articles 1 LEC and 1 SEG. We think that the first paragraph of Article 1 SEG does not provide enough clarity. In light of the above we suggest that “collateral collected as initial margin shall be segregated from proprietary assets on the books and records of the collecting counterparty, a third party holder or custodian, or via other legally binding arrangements made by the collecting counterparty to protect the initial margin from the default or insolvency of the collecting counterparty, third party holder or custodian.”
Finally we believe the requirement for re-investment of cash IM, if maintained, shall exempt the “incidental cash” such as distributions, redemptions and other cash generated by the re-investment).
- Relation between segregation requirements and collateral management requirements (Art. 1 SEG and Art. 2 LEC)
Art. 2 LEC appears to address some aspect which are also covered by the segregation requirements. One example is the requirement under Art. 2 (1) (c) LEC regarding insolvency or bankruptcy remoteness of initial margin maintained with the collateral provider. It is not entirely clear how these obligations covering similar or connected questions are to interact or which requirements prevail in the case of conflict. In particular all requirements addressing insolvency/bankruptcy remoteness need to be coordinated to avoid discrepancies or uncertainties. To this end, these should only be covered by one single provision.
- Assessment of the effectiveness of the netting and segregation agreements
As to the need to for alternative approaches and deviations from the obligation to exchange collateral in the case that segregation agreement and/or netting agreements are not sufficiently protected and supported by the laws of a third country, see our comments regarding this issue in our response to Q1 and Q5 above.
Does this approach address the concerns on the use of cash for initial margin?
- Re-investment of cash initial margin (Art. 1 REU (2))The reinvestment solution in Article 1 REU (2) may not be workable from a practical/commercial perspective. It presumes existing of a framework whereby the custodians would conduct the reinvestment according to the strategy agreed with the parties and moreover in compliance with the credit quality and concentration limits requirements does not currently exists. This therefore does not seem satisfactory to facilitate the usage of cash as IM to the extent it cannot be posted without being re-invested.
The proposal is also too restrictive for other reasons. Indeed, the cash re-used must be reinvested “only for the purpose of protecting the collateral poster” in case of default of the collateral beneficiary. For such purpose, ESAs specify that the cash used must be reinvested in securities and that the latter have to be segregated and not re-used. This restricted discretion granted to the collateral beneficiary may challenge the interest of such re-use while a huge portion of the collateral will be posted in cash.
Respondents are invited to comment on the requirements of this section concerning treatment of FX mismatch between collateral and OTC derivatives.
It should be possible to include the FX volatility in the initial margin calculation instead of using a standard haircut. Further, we would ask the ESAs to provide clarity on the meaning of the term “transfer currency”, and “termination currency”. Does it mean base currency and close out currency?The EBF welcomes that no FX haircut will apply where VM is posted in cash. However, the FX haircuts applicable in all other cases creates prohibitive cost or alternatively an important operational burden on the collateral management should the collateral be posted in currency silos. We support the previous request to replace 8% haircut entirely and include currency mismatches in the IM model for those parties that collect IM.
Regarding FX currency mismatch it is inconsistent that the RTS exempt physically settled FX derivatives from the IM requirements in Art. 5 GEN, but at the same time introduce an 8% haircut as mentioned in paragraph 6 and 7 on page 57. Haircut addresses potential future loss on the value of the collateral. Given that the majority of the risk in a transaction typically is constituted by the FX transaction, it seems unnecessary to introduce haircut on collateral. Furthermore, the haircut of 8% ought to be lower for currency pairs with low volatility, e.g. when one currency is pegged to another.