While EBA is asking for further examples of situations of market stress, we would like to point out, that EBA’s proposal might be based to some extend on a misunderstanding of MiFID II and of Delegated Regulation (EU) 2017/578.
Limited scope of Delegated Regulation (EU) No. 2017/578
EBA states: “The MiFID allows an IF that wishes to operate as market makers on regulated markets and other trading venues (MTF and OTF) to benefit from certain incentives, in exchange for which the IF has to agree to a market making agreement. The Delegated Regulation (EU) No. 2017/578 sets out the detailed obligation for IFs to enter into such a market making agreement and its content as well as obligations upon trading venues for having market making schemes in place.” (Footnote: EBA/CP/2020/06, p. 22, para. 84)
However, this not fully correct. Delegated Regulation (EU) No. 2017/578 does not stipulate a general incentive framework for market makers, in fact it is not even addressed at market makers as such as defined in MiFID II, Article 4 para. 1 (7) (Footnote: Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU ) or in other parts of EU legislation, e.g. in the Short Selling Regulation. (Footnote: Regulation (EU) No 236/2012 of the European Parliament and of the Council of 14 March 2012 on short selling and certain aspects of credit default swaps ) Delegated Regulation (EU) No. 2017/578 rather specifies requirements for an investment firm “that engages in algorithmic trading to pursue a market making strategy" in accordance with and within the narrow scope of Article 17 para. 3 MiFID II. Here, the legislative intent was to implement regulatory minimum standards for “de facto” market makers, (Footnote: Sometimes also referred to as “shadow market makers”.) whose trading strategy was not subject to any regulatory requirements before MiFID II. (Footnote: Recital 60 of MiFID II clarifies „Investment firms that engage in algorithmic trading pursuing a market making strategy should have in place appropriate systems and controls for that activity. Such an activity should be understood in a way specific to its context and purpose. The definition of such an activity is therefore independent from definitions such as that of ‘market making activities’ in Regulation (EU) No 236/2012 of the European Parliament and of the Council.”)
Furthermore, market making agreements and –schemes in the sense of Delegated Regulation (EU) No. 2017/578 are not only restricted to algorithmic trading but also exclusively apply to the market model of “continuous auction order book trading”. (Footnote: Delegated Regulation (EU) No. 2017/578, article 1 para. 1) Consequently, the provisions of Delegated Regulation (EU) No. 2017/578 are not directly relevant for any other market model as defined in MiFID II, RTS 1, Annex 1 and MiFID II, RTS 2, Annex 1.
Since there can be no doubt that the adjustment of the coefficient for K-DTF in the case of stressed market conditions shall be applicable independently from the market model and the trading technology applied, it is more than unfortunate that article 15 (5) of Regulation (EU) 2019/2033 expressly refers to Delegated Regulation (EU) No. 2017/578 and it is paramount to clarify that the definition of “stressed market conditions” as referred to in article 6 of Delegated Regulation (EU) No. 2017/578 in an IFR context should be understood in a generalized way and in particular should not be limited to continuous auction order book trading systems. Accordingly “parameters to identify stressed market conditions in terms of significant short- term changes of price and volume” (Footnote: Delegated Regulation (EU) No. 2017/578, article 6 para. 2) should be regarded to be independent of the market model and the trading venue.
“Stressed market conditions” vs “exceptional circumstances” and “extreme volatility”
While the question of applicable scope of “stressed market conditions” result from a reference in the “Level I” text, we think EBA might also have faced a problem of comprehension when drafting the RTS.
EBA points out: “Article 3 of Delegated Regulation (EU) No. 2017/578 describes ‘exceptional circumstances’ where the obligation for investment firms to provide liquidity on a regular and predictable basis set out in the MIFID shall not apply. In particular, point (a) of Article 3 covers definition of extreme volatility: ‘a situation of extreme volatility triggering volatility mechanisms for the majority of financial instruments or underlyings of financial instruments traded on a trading segment within the trading venue in relation to which the obligation to sign a market making agreement applies.’ Point (a) of Article 3 therefore seems to regard such an extreme volatility situation the circumstances that might potentially be of more relevance to the calculation of the K-DTF.” (Footnote: EBA/CP/2020/06, p. 22, para. 85)
Based on this assessment, EBA suggests: “For the purposes of assessing whether any adjustment to the calculation of K-DTF is required, it seems that only point (a), i.e. a situation of extreme volatility (that triggers volatility mechanisms for the majority of financial instruments on a trading segment within the trading venue) is of relevance.” (Footnote: Ibid para 86)
We think that there is a fundamental misunderstanding here since there is a substantial difference between “stressed market conditions” as stipulated by article 6 (2) of Delegated Regulation (EU) No. 2017/578 (which can apply to a single financial instrument) and “exceptional circumstances” in form of “extreme volatility triggering volatility mechanisms for the majority of financial instruments or underlyings of financial instruments traded on a trading segment within the trading venue” as stipulated by article 3 (a) 2 of Delegated Regulation (EU) No. 2017/578. According to article 15 (5) of Regulation (EU) 2019/2033, a situation of “stressed market condition” should be sufficient to trigger an adjustment of the K-DTF coefficients. However, apart from the general problem described above that the scope of Delegated Regulation (EU) No. 2017/578 is too narrow, EBA’s mandate is limited to find a solution for the adjustment of the coefficients for K-DTF based on the definition of article 6 (2) of Delegated Regulation (EU) No. 2017/578 alone.
Finally and just for the sake of clarity, the application of article 3 (a) of Delegated Regulation (EU) No. 2017/578 would require a situation where more than fifty percent of all financial instruments traded on a specific trading venue or segment of a trading venue would have triggered a volatility mechanism. Actually, we are not aware of a single case where this happened within the Union since Delegated Regulation (EU) No. 2017/578 entered into force. This empirical evidence makes it clear that a situation of “extreme volatility” as described in article 3 (a) 2 of Delegated Regulation (EU) No. 2017/578 was never intended to be the trigger event for an adjustment for the coefficients for K-DTF. It also would be a pretty useless exercise to use this very specific and narrow definition of “extreme volatility” as a trigger event in the K-DTF context since it can be reasonably assumed that any temporary incentives with respect to capital requirements would not stimulate additional liquidity provision in a meaningful way under – fortunately rather theoretical – conditions when quoting obligations are waived and more than fifty percent of all financial instruments traded have already triggered a volatility mechanism.
Guidance on “DTFexcl” and “DTFincl” calculation
To conclude our comments to this question, we think it would be helpful to offer additional explanatory guidance for the definition of “DTFexcl” and “DTFincl” as proposed in draft RTS 9 article 1 (1) (Footnote: EBA/CP/2020/06, p. 63 & 64) since the concept behind the formulas which include derivatives traded in the case of calculating K-DTF for cash trades and cash trades in the case of calculating K-DTF for derivatives trades might not be immediately clear to everyone.
While K-NPR and K-CMG are alternative methods of regulatory capital calculation within the IFR-framework, their original purpose is completely different.
Comparing K-NPR and K-CMG proverbial means comparing “apples and pears”. Aside from the fact that the outcome of the calculations are monetary values which can be comparably higher or lower in a given situation, there is no meaningful criterion – regardless of which value is higher – to decide about which approach is more adequate from a regulatory point of view. K-NPR and K-CMG are simply measuring completely different things.
In this context, it must be remembered that to the extent that K-NPR is considered to be a “risk to market” K-factor, it is simply a fatal methodological misconception. K-NPR, which applies calculations based on CRR methodology, captures “market risk” which is something fundamentally different. From the perspective of the K-factor categorisation, K-NPR is falsely regarded to be “risk to market”, while in fact it is “risk to firm”, arising from assets held whose valuations are subject to fluctuations in market prices. Accordingly, K-NPR measures risks of a firm caused by the markets and not vice versa.
Conversely, K-CMG addresses “risk to market” to the extent that it mitigates settlement risk and that the clearer who requires the collateral is a market participant himself. However, it also should be remembered that it is a core functionality of the clearer to absorb the settlement risk of its clients and to guarantee the fulfilment of the settlement to the client’s market counterpart, no matter whether the client who undertook the trade fails or not.
It is also worth noticing, that while the calculation of K-NPR is based on end of day positions, K-CMG depends on a firm’s trading volume. Accordingly, both values might develop differently over time. In regard of the strong growth in trading activity during the recent month, our observation is that margin requirements have multiplied and grown much faster than NPR/CRR capital requirements.
Accordingly, we are of the opinion that the thresholds which would trigger a comparison of K-CMG to K-NPR proposed in article 4 para. 2 (b) of draft RTS 10 (20% change in K-CMG capital requirements resulting from a change in business strategy of a trading desk and 10 % change in K-CMG capital requirements resulting from a change in the clearing members margin model) are calibrated much too low.
Furthermore, there is no convincing argument that a mandatory comparison between K-CMG and K-NPR should be triggered in cases where K-CMR capital requirements increase. Therefore, we are strongly of the opinion that only decreasing K-CMR capital requirements resulting from changes in business strategy (and not from general market activity) or changes in the clearing member’s margin model should be regarded as trigger events for a mandatory comparison of K-CMG and K-NPR.
Avoiding unnecessary administrative burden
It should be further noted that licencing fees for software which automatically calculates K-NPR/CRR capital requirements and other IT-resources employed for this purpose are a significant cost block in particular for small and mid-sized investment firms. Therefore, it would be clearly disproportionate and in contradiction with the intended avoidance of unnecessary compliance costs and administrative burden by the new framework, (Footnote: Cf. European Commission Proposal for a REGULATION OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL on the prudential requirements of investment firms and amending Regulations (EU) No 575/2013, (EU) No 600/2014 and (EU) No 1093/2010, COM(2017) 790 final, 2017/0359 (COD), Brussels, 20.12.2017, p. 6) if firms would be obliged to employ the resources required to be able to calculate K-NPR and K-CMG parallel at all times. Accordingly if a comparable K-NPR calculation should be triggered for an investment firm which currently uses the K-CMG approach, such a calculation should be allowed to be done in the form of a rough calculation (by using a standard spread-sheet application) first and only if the results would raise regulatory concern, a detailed calculation would be required.
Avoiding undue disadvantages for investment firms using multiple clearers
EBA proposes in Article 3 of draft RTS 10 that investment firms which make use of more than one clearing member “shall calculate the K-CMG by first determining the third highest amount of total margins required on a daily basis by each clearing member separately over the preceding three months, then adding those amounts and multiplying the outcome by 1.3.” (Footnote: EBA/CP/2020/06, p. 69)
The proposed calculation, which sums up the three highest amounts of each clearing member, instead of determining the total margins required across all clearing members an investment firm makes use of first, would put firms with multiple clearing members at an unjustified systematic disadvantage. While it is often not feasible, especially for smaller and mid-size investment firms to use multiple clearers (in particular because this inevitably leads to a fragmentation of collateral), using multiple clearers would reduce the concentration of counterparty risk which the clearer represents from an investment firm’s perspective. Accordingly, it should be desirable from a regulatory point of view (even though it is not feasible for many firms), if an investment firm is able to make use of more than one clearer and it should be rather incentivized than penalized by comparably disadvantageous capital requirement calculations.
Incommensurability of K-NPR and K-CMG
Since K-NPR and K-CMG – as demonstrated in our answer to question 4 – are simply “two different pairs of shoes”, to require an investment firm to “adequately justify the difference between these capital requirements” (Footnote: Ibid, p. 68, draft RTS 10, recital 5), as proposed by EBA, is a meaningless and objectively unachievable task. Furthermore, while it is appropriate and comprehensible to require consistency of application of K-CMG with respect to all positions of a trading desk (Footnote: Cf. EBA/CP/2020/06, p. 68, draft RTS 10, article 4 para. 1 (a)) as well as across trading desks which are “similar in terms of business strategy and trading book positions” (Footnote: CF. Ibid, (b)), it is again an impossible task to require that “K-CMG would be appropriately reflecting the risks of an investment firm’s trading book positions, including expected holding periods” (Footnote: Ibid (c)) simply because K-CMG does not measure position risk, let alone expected holding periods in any way. Therefore, article 4 (1) (c) in draft RTS 10 should be deleted, in order to avoid another fundamental methodological misconception within the IFR-framework.
Switching from K-CMG to K-NPR can never raise concerns of regulatory arbitrage
Article 4 (2) (a) of draft RTS 10 suggests that unless business strategy or operations of a group of dealers has changed, an investment firm should be bound to the use of the K-CMG approach for a period of at least 24 months. (Footnote: Cf. EBA/CP/2020/06, p. 70) We think that this restriction – in its generalisation – lacks a convincing rationale.
In this context, it is worthwhile remembering that – as already mentioned in our answer to question 4 – despite of the CRR based K-NPR is a methodological misconception in so far as it does not capture “risk to market”, it needs to be recognized that it calculates capital requirements for trading book activities in a way which is deemed sufficient and appropriate for systemically important “Class 1” firms and credit institutions which fund their trading book activities at least in part by deposits. (Footnote: Even though it does not require investment firms to adopt recent and future changes to CRR.) Therefore, if a firm voluntarily wishes to switch from K-CMG to K-NPR, it can hardly be regarded as regulatory arbitrage, even if the change should result in a decrease of capital requirements. Accordingly, investment firms should be enabled to switch from K-CMG to K-NPR at any time. However, we understand that it is not desirable from a regulatory point of view, if firms change the way of calculating their capital requirements too often. We therefore suggest, that the proposed 24 months period (Footnote: Under the conditions laid down in draft RTS 10, article 4 para. 2 (a), EBA/CP/2020/06, p.70) should be applied only if a firm who has switched from K-CMG to K-NPR before wishes to re-implement the K-CMG approach.
With regard to the draft RTS 5 on prudential consolidation of investment firm groups (Article 7(5) of the IFR) we share the same opinion with our colleagues of BVI Bundesverband Investment und Asset Management, whose comments we adopt:
“In general, we strongly disagree with the approach taken by the EBA in defining a completely new scope of group constellations in Articles 2 to 5 of the Draft RTS. Such an extension of the scope is not covered by the mandate given in Article 7(5) IFR which states that the EBA shall develop draft RTS to specify ‘the details of the scope and methods for prudential consolidation of an investment firm group, in particular for the purpose of calculating the fixed overheads requirement, the permanent minimum capital requirement, the K-factor requirement on the basis of the consolidated situation of the investment firm group, and the method and necessary details to properly implement paragraph 2’ of Article 7 IFR. That mandate limits the EBA to develop details on the scope for prudential consolidation within the given definitions and requirements in the IFR and not to develop completely new principles of the scope.
Moreover, the new proposals in Articles 2 to 5 of the Draft RTS are in considerable contradiction to the approach taken by the IFR definition of an investment firm group with reference to Article 22 of Directive 2013/34/EU. In particular, the cases defined in Article 22 of that Directive would be undermined by the proposed Articles 2 to 5 of the Draft RTS. In addition, Articles 2 to 5 of the Draft RTS considerably deviate from the current regulations on own funds on a consolidated basis for groups consisting of investment firms only (i.e. without any credit institutions) according to Article 98 CRR II. This is not in line with the purpose described in Recital 12 of the IFR to mirror the existing treatment of such investment firm groups under the CRR and CRD. The EBA itself states the need to ensure such a consistency in Recitals 3 and 4 of the Draft RTS. In this context, it is not appropriate to copy a draft RTS established under the CRR in 2017 with divergent legal basis that did not enter into force - also due to the justified criticism of the banking industry.
Furthermore, according to Article 7(2) and Recital 12 of the IFR, the parent undertaking of an investment firm group should be required to comply with the requirements of the IFR based on the consolidated situation of the group. We therefore strongly disagree with defining new responsibilities such as that an investment firm being a subsidiary in an investment firm group (for instance as part of a holding structure) should ensure that other entities within the group that are not subject of the IFR implement arrangements, processes and mechanisms to ensure proper consolidation. That would lead to the situation that such an (as the case may be, very small-sized) investment firm needs to be capitalised to fulfil the capital requirements in the group on a consolidated basis.
Hence, we have expected that the Draft RTS will deal with principles regarding consolidation methods considering the special features of management companies licenced under the AIFMD or UCITS Directive, investment funds such as UCITS or AIF or securitisation special purpose entities being part of a group. Statements on this are completely missing in the draft RTS.
More specifically, we suggest the following amendments to the Draft RTS:
1. Article 2(1) of the Draft RTS: Group of undertakings which meet the conditions set out in Article 22 of Directive 2013/34/EU
We urge the EBA to delete Article 2(1) of the Draft RTS which refers to group constellations of paragraph 7 of Article 22 of Directive 2013/34/EU. Such an approach would be not in line with the definition set out in Article 4(1)(25) IFR. According to that definition, a group of undertakings which meets all the conditions set out in Article 22 of Directive 2013/34/EU (and not limited to those of paragraph 7) should be qualified as an investment firm group. That involves cases where Member States are required to draw up consolidated financial statements and a consolidated management report if a parent undertaking fulfils certain conditions such as it has a majority of the shareholders' or members' voting rights in a subsidiary undertaking.
Moreover, and much more importantly, the approach proposed by the EBA in Article 2 of the Draft RTS would ignore that the conditions in paragraph 7 depend on the implementation by the Member state in its national law because according to Article 22(7) of Directive 2013/34/EU, a Member State may require any undertaking governed by its national law to draw up consolidated financial statements and a consolidated management report if certain conditions are fulfilled (such as non-related undertakings are managed on a unified basis in accordance with a contract). Therefore, the EBA approach set out in Article 2 of the Draft RTS would lead to the situation that these undertakings would be qualified as an investment firm group which are required to comply with certain rules of the IFD and IFR on a consolidated basis in any case. That is much stricter as required under the IFR and should be considered.
2. Articles 3 to 5 of the Draft RTS: Extending the definition of an investment firm group
We urge the EBA to delete Articles 3 to 5 of the Draft RTS. We strongly disagree with the proposed substantial extension of the scope by adding further group constellations such as undertakings with significant influence without participation or capital ties (Article 3 of the Draft RTS), single management other than pursuant to a contract, clauses in memoranda or articles of association (Article 4 of the Draft RTS) or participations or capital ties (Article 5 of the Draft RTS).
We are aware that in its hearing on 30 June 2020, the EBA referred to its consultation on technical standards specifying the methods of prudential consolidation under Article 18 of Regulation (EU) No 575/2013 (Capital Requirements Regulation - CRR), 9 November 2017. (Footnote: Available under the following link: https://eba.europa.eu/sites/default/documents/files/documents/10180/2019694/3b8e5188-f7e3-4d11-b9ae-256e47d61e4b/Consultation%20Paper%20on%20RTS%20on%20methods%20of%20prudential%20consolidation%20%28EBA-CP-2017-20%29.pdf) However, this ignores the fact that the legal bases regarding the scope are fundamentally different in the IFR on the one hand and in the CRR on the other and do not justify such an extension. In detail:
- egal definition of an ‘investment firm group’: Article 4(1)(25) IFR defines for the first time the term of an ‘investment firm group’ with the following group constellation of which at least one is an investment firm and which does not include a credit institution:
- a group of undertakings which consists of a parent undertaking and its subsidiaries or a group of undertakings which meet the conditions set out in Article 22 of Directive 2013/34/EU of the European Parliament and of the Council.
According to this clear definition, there is no room for extending the scope of the definition of an investment firm group to further cases as proposed by the EBA in Articles 3 to 5 of the Draft RTS. This is a major difference compared to the existing rules on prudential consolidation under the CRR because the CRR does not even define the term of an ‘investment firm group’. Any such extension would require respective Level 1 amendments of IFR and cannot be effected by means of RTS.
- The IFR does not contain a provision which is similar with Article 18(6) CRR: According to Article 18(6) CRR, competent authorities shall determine whether consolidation is required in case of significant influence without a participation or other capital ties and single management other than pursuant a contract, memorandum or articles of association. There is neither a comparable regulation in the IFR nor a mandate to specify such cases in a Draft RTS under the IFR. The reference to a similar approach of the EBA public consultation on technical standards specifying the methods of prudential consolidation under Article 18 CRR is therefore in no way appropriate. This applies even more as the RTS on methods of prudential consolidation under Article 18 CRR is under development and has not even entered into force yet. (Footnote : Cf. the reference made by the EBA itself on its website: https://eba.europa.eu/regulation-and-policy/accounting-and-auditing/rts-on-methods-of-prudential-consolidation)
- Articles 3 to 5 of the Draft RTS are much more stringent as the approach of Article 18(6) CRR: We understand the proposed Articles 3 to 5 in conjunction with Articles 6 to 7 of the Draft RTS in such a way that competent authorities should only have a right to choose the consolidation method (such as full or proportional consolidation, aggregation method), but no longer whether consolidation is required or not. Irrespective of the lack of a legal basis for such an approach, this would be much more stringent as it is currently required under the CRR: according to Article 18(6) CRR, competent authorities shall determine whether consolidation is required in additional cases such as of significant influence or single management.
3. Contractual arrangements: delegation of functions in the asset management sector
Irrespective of the legal basis and the substantial extension of the scope of the prudential consolidation of an investment firm group, the cases proposed by the EBA in Articles 2 and 3 of the Draft RTS would lead to inappropriate group constellations in the asset management sector. Because asset management companies qualify as financial institutions in the meaning of the definition in Article 4(1)(14) IFR, they could be related with an investment firm on a contractual basis and would fulfil the proposed group approach proposed by the EBA.
The following example will demonstrate this:
- According to the UCITS Directive and the AIFMD, management companies can delegate portfolio management services of the investment funds such as UCITS or AIF to third parties (such as investment firms with a licence to provide asset management) on a contractual basis. That case is fully covered by the prudential requirements of the AIFMD or UCITS Directive. However, Articles 2 and 3 of the Draft RTS on prudential consolidation of investment firm groups could be understood in such a way that such a contract would qualify as a significant influence without participation or capital ties. This would lead to the situation that the investment firm and the management company would be qualified as an investment firm group with the effect that the investment firm must carry out consolidation of the management company although this case is already comprehensively covered by the UCITS Directive or the AIFMD.
In the further alternative, if EBA maintains its approach, we urge to clarify that cases where an AIF or UCITS management company delegates functions such as the portfolio management to an investment firm on a contractual basis are out of scope of prudential consolidation.”