Response to second Joint Consultation on draft RTS on risk-mitigation techniques for OTC-derivative contracts not cleared by a CCP
Go back
Article 1 VM par. 1 (page 31): DACSI believes that minimum transfer amounts should apply.
Article 1 GEN par. 4 (page 26): DACSI notes that the definition of initial margin is incorrect: the objective of initial margin is not providing protection against current exposure; this is provided by variation margin (based on the valuation at close-of-business on the previous business day). The reference to “current exposure” should consequently be removed from this article.
In line with the comment above, there is an inconsistency between Article 1.1 VM (page 31) and Article 1.5 GEN (page 27), i.e. the latter explicitly refers to the “daily marking-to-market”, which should be the principle.
Minimum transfer amounts should apply to the initial margin and variation margin separately i.e. leading to one minimum transfer amount for initial margin and one minimum transfer amount for variation margin. This principle does not appear in the RTS but solely in the considerations (recital 10, page 19). For legal certainty this principle should be provided for in the RTS.
Article 1 VM par. 5 (page 32) versus Article 1 VM par. 3 (page 31): DACSI strongly believes that it is not possible to monitor different rules in respect of settlement timing (one business day rather than 3 business days) depending on the fact whether or not initial margin requirements apply. This would mean a costly and lengthy IT setup which is undesirable. In addition for certain markets one business day is not possible (for example Japanese market). Consequently a consistent and uniform approach regarding settlement period for posting of variation margin i.e. not dependent on whether or not initial margin is exchanged, should be adopted. In addition, the settlement period should be dependent on the local markets (variation between 1 to 3 business days).
Article 1 EIM (page 32)/Articles 1 MRM to 6 MRM: Initial margin models: As the RTS is currently structured, the likelihood of differences occurring in model outcomes across two counterparties each using their internal models is very high. It would likely result in an unworkable situation with a third party model or the standardised approach becoming the only viable alternatives. The ISDA SIMM model would be an acceptable model as well. It is proposed to use harmonised and standardised models. This would ensure consistency and legal certainty, and would avoid lengthy discussions between parties with the risk that the big/powerful parties will impose their own models.
Article 1 MRM (page 33): if differences exist it will become very difficult to manage portfolios. DACSI sees that for example in the BCVA discussions. The difference in framework between RC CVA and BCVA (accounting), managing RC CVA by buying credit protection (CDSs) can result in large accounting losses. It also makes models difficult to maintain.
Article 3 MRM (page 34): creating one data set with non-stressed data and adding stressed data will result in non-observed historical market data e.g. volatilities and correlations. This will influence volatilities and also correlations. Stressed periods have higher correlations between risk drivers than in non-stressed markets. The preference is to have a consistent calibration over observed market data. In CRD IV they do not commingle market data but uses two calibration files one for the current period and one for the stressed period.
Article 3 MRM (page 34): DACSI anticipates that the specified frequency by which back-testing should be performed will not add significant value as only a limited sample of the data would be replaced on a quarterly basis (approx. 5%), if that part of the time series is included at all, and not substituted by historic stress data. A frequency for back testing of 6 months would be more appropriate to ensure there is a materially different time series against which calibration is performed.
Article 2 OPD par.2 (page 48): DACSI suggests a closer alignment with the existing requirement under Article 296 of the Capital Requirements Regulation. Regulated entities are already required to perform legal review by use of legal opinions.
DACSI strongly opposes any discrepancy between conditions on contractual agreements in general and on re-use of collateral in particular set in different European regulations, as this would unnecessarily restrict economic activity and create legal uncertainty about existing and future transactions. We firmly propose to align the provisions on re-use in the RTS, in the SFT Regulation and in any other regulation.
DACSI is convinced that the requirement for the protection of cash in case of insolvency of the custodian is not realisable: this is not possible even in the UK (no client money protection) because cash is not protected in general in such an event (cash deposited with custodians does not benefit from the client money protection). In addition even if securities are exchanged as initial margin there will always be some cash components (dividends/coupons). A solution is to be found in the protection of cash through the RTS when deposited with a custodian in the European Union or to open the possibility to deposit the cash with central banks. If no solutions are found, this will de facto mean that no cash can be exchanged as initial margin because no segregation/protection in case of insolvency is possible in all European Union jurisdictions.
Respondents are invited to comment on the proposal in this section concerning the timing of calculation, call and delivery of initial and variation margins.
Article 1 VM par. 1 (page 31): DACSI thinks that the valuation based on the current valuation is impossible from a monitoring/operational point of view because this would involve a continuous monitoring whereas collateral is exchanged/posted generally once a day. The valuation should be based on close-of-business on the previous business day, which is standard market practice. In addition, please note that the haircuts and initial margin requirements will ensure for a protection against volatility of the instruments given as collateral and the underlying of the OTC derivatives which would suffice to build up buffers.Article 1 VM par. 1 (page 31): DACSI believes that minimum transfer amounts should apply.
Article 1 GEN par. 4 (page 26): DACSI notes that the definition of initial margin is incorrect: the objective of initial margin is not providing protection against current exposure; this is provided by variation margin (based on the valuation at close-of-business on the previous business day). The reference to “current exposure” should consequently be removed from this article.
In line with the comment above, there is an inconsistency between Article 1.1 VM (page 31) and Article 1.5 GEN (page 27), i.e. the latter explicitly refers to the “daily marking-to-market”, which should be the principle.
Minimum transfer amounts should apply to the initial margin and variation margin separately i.e. leading to one minimum transfer amount for initial margin and one minimum transfer amount for variation margin. This principle does not appear in the RTS but solely in the considerations (recital 10, page 19). For legal certainty this principle should be provided for in the RTS.
Article 1 VM par. 5 (page 32) versus Article 1 VM par. 3 (page 31): DACSI strongly believes that it is not possible to monitor different rules in respect of settlement timing (one business day rather than 3 business days) depending on the fact whether or not initial margin requirements apply. This would mean a costly and lengthy IT setup which is undesirable. In addition for certain markets one business day is not possible (for example Japanese market). Consequently a consistent and uniform approach regarding settlement period for posting of variation margin i.e. not dependent on whether or not initial margin is exchanged, should be adopted. In addition, the settlement period should be dependent on the local markets (variation between 1 to 3 business days).
Article 1 EIM (page 32)/Articles 1 MRM to 6 MRM: Initial margin models: As the RTS is currently structured, the likelihood of differences occurring in model outcomes across two counterparties each using their internal models is very high. It would likely result in an unworkable situation with a third party model or the standardised approach becoming the only viable alternatives. The ISDA SIMM model would be an acceptable model as well. It is proposed to use harmonised and standardised models. This would ensure consistency and legal certainty, and would avoid lengthy discussions between parties with the risk that the big/powerful parties will impose their own models.
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.
Article 6 MRM par. 1(a) (page 37): there is no definition of “suitably qualified and independent parties”. This should be defined in the RTS to provide for legal certainty and consistency.Article 1 MRM (page 33): if differences exist it will become very difficult to manage portfolios. DACSI sees that for example in the BCVA discussions. The difference in framework between RC CVA and BCVA (accounting), managing RC CVA by buying credit protection (CDSs) can result in large accounting losses. It also makes models difficult to maintain.
Article 3 MRM (page 34): creating one data set with non-stressed data and adding stressed data will result in non-observed historical market data e.g. volatilities and correlations. This will influence volatilities and also correlations. Stressed periods have higher correlations between risk drivers than in non-stressed markets. The preference is to have a consistent calibration over observed market data. In CRD IV they do not commingle market data but uses two calibration files one for the current period and one for the stressed period.
Article 3 MRM (page 34): DACSI anticipates that the specified frequency by which back-testing should be performed will not add significant value as only a limited sample of the data would be replaced on a quarterly basis (approx. 5%), if that part of the time series is included at all, and not substituted by historic stress data. A frequency for back testing of 6 months would be more appropriate to ensure there is a materially different time series against which calibration is performed.
Respondents are invited to comment on whether the requirements of this section concerning the concentration limits address the concerns expressed on the previous proposal.
Article 7 LEC: the limitation imposed does already have a threshold. Note that both these limitations will have an impact on counterparties’ investment policies, possibly resulting in a liquidity squeeze in the bond market, and to a lesser extent in the equity and commodity markets. Preferable and more effective are general limitations on acceptable percentage of issue size. This also serves the purpose of ensuring liquidity in case of counterparty’s default.Respondent to this consultation are invited to highlight their concerns on the requirements on trading relationship documentation.
Article 2 OPD par. 1 (page 47) in combination with Article 1 FP par.2 (page 52): non-EEA NFC- counterparties do not fall under the scope of the margin requirements and this is the appropriate approach to ensure for a global level playing field and a non-disadvantageous position for EU parties. However, there are now some additional documentation requirements for all parties (FC, NFC+, NFC-, EEA/non-EEA, intra-group and for all products) that lead to the obligation to sign ISDA documentation (or other equivalent documentation) in all cases. This is of course acceptable. However, the timing is supposed to be ’immediate’. Some more time is needed for parties to comply with these requirements since such a repapering exercise will take substantial time (lengthy negotiations). DACSI proposes to align the timing of this requirement with the variation margin requirements in order for the relevant parties to have sufficient time to document the OTC derivative transactions.Article 2 OPD par.2 (page 48): DACSI suggests a closer alignment with the existing requirement under Article 296 of the Capital Requirements Regulation. Regulated entities are already required to perform legal review by use of legal opinions.
Respondents are invited to comment on the requirements of this section concerning the legal basis for the compliance.
N/ADoes this approach address the concerns on the use of cash for initial margin?
Article 1 REU par. 1 (page 49): this article prohibits re-hype/re-use of initial margin. This should not apply when title transfer is concerned: title transfer by nature and by definition provides the collateral giver’s consent to the collateral taker to re-hypothecate or otherwise re-use the collateral. We refer to the lengthy considerations spent on this issue in the current preparation of the SFT Regulation and to the most recent wording in the draft re Article 15 (Reuse of financial instruments received under a collateral arrangement). This article specifies conditions on re-use, but does not prohibit it.DACSI strongly opposes any discrepancy between conditions on contractual agreements in general and on re-use of collateral in particular set in different European regulations, as this would unnecessarily restrict economic activity and create legal uncertainty about existing and future transactions. We firmly propose to align the provisions on re-use in the RTS, in the SFT Regulation and in any other regulation.
DACSI is convinced that the requirement for the protection of cash in case of insolvency of the custodian is not realisable: this is not possible even in the UK (no client money protection) because cash is not protected in general in such an event (cash deposited with custodians does not benefit from the client money protection). In addition even if securities are exchanged as initial margin there will always be some cash components (dividends/coupons). A solution is to be found in the protection of cash through the RTS when deposited with a custodian in the European Union or to open the possibility to deposit the cash with central banks. If no solutions are found, this will de facto mean that no cash can be exchanged as initial margin because no segregation/protection in case of insolvency is possible in all European Union jurisdictions.