Response to discussion on the potential review of the investment firms’ prudential framework

Q1: What would be the operational constraints of potentially removing the threshold?

We believe this question is best answered by firms themselves, as they are directly impacted by such regulatory thresholds and are best positioned to provide detailed insights into the operational constraints they might face.

Q2: Would you suggest any further element to be considered regarding the thresholds used for the categorisation of Class 3 investment firms?

We do not suggest adding any further elements to the list of conditions for qualifying as a Class 3 firm. However, from a methodological perspective, we propose considering the elimination of conditions (h) and potentially reducing the quantitative thresholds in conditions (a) and/or (b).

Due to the cumulative nature of art. 12 IFR, conditions (h) and (i) only practically serve to classify certain firms as Class 2 instead of Class 3. Specifically, these are firms whose:

1. ASA, CMH, DTF, NPR/CMG and TCD are zero;
2. AUM is less than EUR 1.2 billion;
3. COH is less than EUR 100 million/day for cash trades and EUR 1 billion/day for derivatives,

but whose:

1. balance-sheet total is more than EUR 100 million and/or total annual gross revenues from investment services and activities are more than EUR 30 million.

In all other scenarios where an investment firm qualifies as Class 2 instead of Class 3 due to conditions (h) and/or (i), it does so because it also triggers other conditions under (a) to (g) of art. 12 IFR.

Consequently, in the unique scenario described above, these “large” firms only concern firms providing limited (i) portfolio management, (ii) investment advice, (iii) reception and transmission of client orders, and/or (iv) execution of order services, since their only non-zero elements could be:

• AUM, but less than EUR 1.2 billion; and/or
• COH, but less than EUR 100 million/day for cash trades and EUR 1 billion/day for derivatives.

If such a firm poses a risk warranting Class 2 classification (i.e., meaning it should consider the K-factor requirements for the calculation of its own funds requirements), we believe it would be methodologically more sound to eliminate conditions (h) and (i) and potentially reduce the quantitative thresholds under conditions (a) and/or (b) to a level that warrants Class 2 firm qualification if the conditions are not met. However, we do not see any direct reason for reducing the latter thresholds.

The IFR stipulates that firms should be considered to be Class 3 firms for the purposes of the specific prudential requirements for investment firms where they do not conduct investment services which carry a high risk for clients, markets or themselves and where their size means they are less likely to cause widespread negative impacts for clients and markets if risks inherent in their business materialise or if they fail.[1]

However, in our view, the decisive factor to warrant Class 2 classification should be the nature of the investment services provided by the firm – such as the level of AUM and COH activities - which directly impact client risks and the firm’s operational risk profile. An increased balance sheet size alone does not necessarily reflect the risk posed to clients or the market. These risks are already addressed by the FOR applicable to Class 3 firms, whose prudential purpose is to serve as a capital cushion to ensure an orderly wind-down of a firm. Therefore, firms should be classified as Class 2 based on their increased provision of investment services and corresponding risks, rather than solely due to a larger size, which would inadvertently inflate regulatory burdens without addressing the underlying risk factors.

This approach would also align the conditions for Class 3 classification (i.e., staying below the AUM and COH thresholds) with the consequences of failing to meet these conditions (i.e., facing higher own funds requirements resulting from the switch to the K-factor regime and having a high K-AUM and/or K-COH calculation). After all, a mere increase in balance size does not warrant calculation of K-factors (nor does it even necessarily increase K-factors), nor adds to prudency, if they remain zero or within the preset limits.

Therefore, we propose eliminating conditions (h) and (i) completely and decreasing the quantitative elements in conditions (a) and (b) to a level that warrants Class 2 firm qualification if the conditions are not met.

[1]            Recital (17) IFR.

Q3: Do you have any views on the possible ways forward discussed above regarding the transition of investment firms between Class 2 and Class 3 should be introduced?

We believe that frequent reclassification of investment firms within the same financial year or over a 12-months period is undesirable as it causes uncertainty for such firms. Accordingly, we propose that, once a firm no longer meets the conditions to qualify as a Class 3 firm, it shall be a Class 2 firm for a period of at least 12 months.

To address these issues, we suggest amending art. 12(3) IFR as follows:

“3. Where an investment firm no longer meets all the conditions set out in paragraph 1, it shall cease to be a small and non‐interconnected investment firm for a period of 12 months with immediate effect.

By way of derogation from the first subparagraph, where an investment firm no longer meets the conditions set out in points (a), (b), (h) or (i) of paragraph 1 but continues to meet the conditions set out in points (c) to (g) of that paragraph, it shall cease to be considered to be a small and non‐interconnected investment firm after a period of three months and for a period of at least 12 months, calculated from the date on which the threshold was exceeded. The investment firm shall notify the competent authority without undue delay of any breach of a threshold.”

Q4: Should the minimum level of the own funds requirements be different depending on the activities performed by investment firms or on firms’ business model? If yes, which elements should be considered in setting such minimum?

While we acknowledge that there may be a case for varying the fixed overheads requirement (“FOR”) based on the activities performed by investment firms or their business models, we ultimately recommend against introducing a more complex FOR resulting in higher upfront Pillar 1 capital requirements.

In favour of an activities-based FOR, we argue the following. If the FOR were to depend on the activities performed or the business model of an investment firm, those that conduct more activities (i.e., generally those with a more challenging wind-down process) would be required to hold more months of fixed overheads as own funds compared to firms with a simpler business model (i.e., those with a generally simpler wind-down process). This could make the FOR a more accurate proxy for facilitating an orderly wind-down. For instance, firms servicing many clients may face more difficulties in ceasing operations compared to those with no clients. However, for the latter firms (i.e., that deal on own account), the FOR no longer primarily addresses operational risks as it did under CRR, but rather acts as a de facto capital floor if the K-Factor requirement is lower, ensuring sufficient capital for an orderly wind-down.[1] Thus, a pure activities-based FOR would have to account purely for the wind-down scenario. In our experience, the size of an investment firm can also be a crucial factor in determining wind-down complexity. A large Class 2 (almost Class 1) firm would generally be harder to wind down than a small Class 3 firm, as it can be expected that it will have generally more, and generally more complex, relations. However, equally so, size does not necessarily express complexity in wind-down, as large firms may be equally easy to wind down when they have a simple business model. For instance, large trading firms that only deal in cleared trades through a clearing member and CCP should theoretically be fairly easily (orderly) wound down regardless of their trading book size. Conversely, a small firm with many client contracts and service agreements with other financial institutions might be fairly hard to wind down due to varying termination periods and rights. Though, these are specific cases that are a deviation of the possible general rule that larger firms are more difficult to wind down than smaller firms. Against this background, we believe there could be scope to argue in favour of calibrating the FOR based on (a) the size of the firm and (b) the investment services or activities conducted.

However, against such an activity- and size-based FOR, we argue that the approach outlined above essentially mirrors a stratified and less granular Pillar 2 assessment of a firm´s wind-down capacity. We note that the SREP Guidelines already address the adequacy of wind-down capital. In our view, such a firm-specific assessment under Pillar 2 is more suitable than any non-granular regime under Pillar 1. For example, the nuance in the size criterion is best addressed on a case-by-case analysis rather than by general rules. The K-factor regime is already designed to cover firm-specific risks related to the services and activities conducted. Introducing an activity-based variation in the FOR might duplicate the K-factor regime and hollow out the Pillar 2 assessment.

Therefore, we suggest not developing a more complicated or higher FOR on an activity based argumentation or on a general extension of the relevant period, without any fundamental argumentation (e.g., if a quantitative survey by EBA and ESMA were to indicate that the average wind-down period for investment firms exceeds three (3) months[2]) as to why such period would need to be longer.

It should be noted however that a higher generic Pillar 1 FOR requirement prevents the tailor made solutions and assessments that are purported to be made in the current regime in the context of the SREP, based on the criteria of Article 1 Delegated Regulation (EU) 2023/1668. A careful alignment of the Level 1 text and Level 2 rules should be conducted to prevent overlap or inconsistencies in approaches.

[1]            Discussion Paper, (2024), par. 47, 49-50; EBA Final Report, On Investment Firms, (2015), p. 60.

[2]            In line with the original reasoning behind the FOR in the IFR, see: EBA Final Report, On Investment Firms, (2015), p. 61.

Q5: Is it necessary to differentiate the deductibles by activity or by business model for the purpose of calculating the FOR? If yes, which items should then be considered and for what reasons?

We believe that, for the same reasons mentioned in our response to Q4, it is not necessary to differentiate the deductibles by activity or business model for calculating the FOR. Introducing such differentiation would complicate the FOR unnecessarily, moving it towards a more complex model akin to the SREP assessment. Including specific business activities as deductible items would shift the FOR from its basic subtractive approach towards a more complicated (almost) additive approach, which is undesirable in our view. We recall that the subtractive approach was chosen over the additive approach as the latter was deemed impractical and less prudent than the former.[1]Therefore, we oppose including further specific deduction grounds in the RTS specifying certain aspects of the FOR (“FOR RTS”)[2] or art. 13(4) IFR.

However, we see merit in introducing a residual general deduction ground in art. 13(4) IFR, which includes certain deduction grounds, with further specifications in the FOR RTS. These current deduction grounds are non-exhaustive, as indicated by EBA during the drafting of the FOR RTS.[3] However, in our experience, competent authorities tend to apply these grounds strictly, generally refusing deductions outside the specified list in art. 13(4) IFR or the FOR RTS. This rigidity is further exacerbated by the IFREP report, which does not clearly allow for a residual category of deduction grounds, forcing firms to appoint residual incidental costs to inappropriate entries, creating data quality issues.

To address this issue, we propose including a residual deduction ground in art. 13(4) IFR. This would allow firms to deduct variable costs not covered by the current deduction grounds. Permissions for deductions on this residual ground could be subject to supervisory approval and guided by specific EBA guidelines. This would prevent regulatory arbitrage and ensure a level playing field across Member States, though it would not stratify as much as direct inclusion in the RTS. Hence, it would provide legal certainty for firms and allow the EBA the flexibility to identify business activities deserving of specific deduction through guidelines, which if deemed necessary could then be implemented in the FOR RTS.

We concretely propose to amend art. 13(4) IFR as follows:

‘4. EBA, in consultation with ESMA, shall develop draft regulatory technical standards to supplement the calculation of the requirement referred to in paragraph 1 which includes at least the following items for deduction: 

(a) staff bonuses and other remuneration, to the extent that they depend on the net profit of the investment firm in the respective year; 

(b) employees’, directors’ and partners’ shares in profits; 

(c) other appropriations of profits and other variable remuneration, to the extent that they are fully discretionary; 

(d) shared commission and fees payable which are directly related to commission and fees receivable, which are included within total revenue, and where the payment of the commission and fees payable is contingent on the actual receipt of the commission and fees receivable; 

(e) fees to tied agents; 

(f) non‐recurring expenses from non‐ordinary activities;

(g) other incidental variable costs.

Competent authorities shall allow an investment firm to deduct other incidental variable costs included in sub-paragraph (g), provided that the investment firm has demonstrated to the competent authority that the item submitted for deduction cannot reasonably be identified as a fixed cost and cannot be allocated to any of the other deduction grounds included in sub-paragraphs (a) to (f) of this paragraph.

EBA shall also specify for the purposes of this Article the notion of a material change. 

EBA shall submit those draft regulatory technical standards to the Commission by 26 December 2020. Power is delegated to the Commission to supplement this Regulation by adopting the regulatory technical standards referred to in the first subparagraph in accordance with Articles 10 to 14 of Regulation (EU) No 1093/2010.

(5) For the purposes of sub-paragraph (g) of paragraph 4, EBA, in consultation with ESMA, shall issue guidelines specifying further the costs that competent authorities may recognise as eligible for deduction as other incidental variable costs in the calculation of the requirement referred to in paragraph 1.’

[1]            EBA, Final Report on own funds requirements for investment firms based on fixed overheads under Article 97(4) of Regulation (EU) No 575/2013 (Capital Requirements Regulation – CRR), (2014), p. 3-4.

[2]            Commission Delegated Regulation (EU) 2022/1455 of 11 April 2022 supplementing Regulation (EU) 2019/2033 of the European Parliament and of the Council with regard to regulatory technical standards for own funds requirement for investment firms based on fixed overheads.

[3]            EBA Final Report, On the draft RTS on the implementation of the IFR/IFD, (2020), p. 8.

Q6: Are expenses related to tied agents material for the calculation of the FOR to the extent to require a dedicated treatment for their calculation? If yes, are the considerations provided above sufficient to cover all the relevant aspects?

We understand that the rationale behind including third-party costs incurred on behalf of the firm, including those related to tied agents, in the fixed expenses of that firm is to prevent firms from outsourcing activities to artificially lower their fixed expenses base.[1] In principle, we agree with this rationale since these costs would still be owed to third parties in a gone concern scenario. As mentioned in the Discussion Paper, the FOR serves as a gone concern capital floor, facilitating an orderly wind-down.

We recall that the prudential purpose of the FOR is to provide a “capital cushion” for an orderly wind-down of a firm. It was estimated that such a wind-down should not take significantly longer than three (3) months, hence the FOR was set at one-fourth (25%) of the fixed overheads over the past financial year. During the initial drafting of the IFR regime, the calibration of the FOR under CRR was deemed appropriate. In cases where a firm might need more than three (3) months, a Pillar 2 add-on was considered more suitable to cover wind-down capital shortcomings than further adjusting the FOR.

The capital retained under the FOR requirements is primarily meant to be used in a wind-down scenario, i.e., the capital serves to pay creditors and prevent insolvency procedures before the firm’s commercial activities can be properly wound down. Therefore, including third-party fixed costs incurred on behalf of the firm should be clearly anchored in a legally enforceable obligation (e.g., an agreement). Without such an obligation, there is no reason to assume that a firm should continue to pay the third party during an orderly wind-down.

If including such costs aims to prevent regulatory arbitrage by firms that might outsource costs without remuneration, we believe the same reasoning applies. A firm that shifts fixed costs to another party without paying fair compensation (e.g., group entities making non-arms length arrangements) would naturally have lower costs. The same reasoning applies to tied agents under art. 29 MiFID II. Firms are not under a regulatory obligation to pay tied agents a fixed amount; the remuneration for services rendered by the tied agents should be clearly detailed in an agreement.[2]

We therefore propose a strict legalistic approach for including third-party costs. If there is no legally enforceable claim, these costs should not be included in the FOR. However, we do not believe that third-party costs should be included when such costs are already covered within the total expenses in the annual financial statements referred to in paragraph 1 as an expense of the investment firm for the payment of a fair compensation for the activities performed by the third party that incurred the fixed expenses.

We therefore recommend amending art. 1(5) FOR RTS as follows:

‘5. Where third parties, including tied agents, incurred fixed expenses, on behalf of the investment firms, under a legally enforceable agreement between the investment firm and the third party,and therefore have not already been included within the total expenses in the annual financial statements referred to in paragraph 1, those fixed expenses shall be added to the total expenses of the investment firm, unless those fixed expenses are already covered within the total expenses in the annual financial statements referred to in paragraph 1 as an expense of the investment firm for the payment of a fair compensation for the activities performed by the third party that incurred the fixed expenses. Where a breakdown of the third party’s expenses is available, an investment firm shall add to the figure representing the total expenses only the share of those fixed expenses applicable to the investment firm. Where such a breakdown is not available, an investment firm shall add to the figure representing the total expenses only its share of the third party’s expenses as it results from the business plan of the investment firm.’

[1]            EBA Discussion Paper, (2016), p. 31.

[2]            ESMA, Supervisory Briefing: On supervisory expectation in relation to firms using tied agents in the MiFID II framework, (2022), p. 8.

Q7: Should the FOR be calculated distinguishing the costs related to non-MiFID activities, which criteria should be considered? What kind of advantages or disadvantages would this have in practice?

In our view, the FOR should not be linked to the type of activities conducted by the firm. We reiterate that the FOR primarily serves as a capital floor to ensure an orderly wind-down of the firm, functioning as gone concern capital. As gone concern capital, it is irrelevant to differentiate between types of activities being wound down, as the entire legal entity must be wound down in such a scenario. The FOR is designed not to address significant sums of going concern capital but rather to ensure sufficient gone concern capital, based on the fundamental principle that investment firms are generally not systemically important and should be allowed to fail.

In instances where the FOR capital is used to absorb costs in an orderly wind down scenario, the process must culminate in the cessation of all commercial activities of the relevant legal entity of the firm.[1] Therefore, differentiating between activities for the purposes of FOR seems rather theoretical, as this would hinder the orderly wind-down of MiFID activities. Ultimately, creditors related to MiFID activities generally rank pari passu with creditors related to other services in insolvency scenarios. Hence, MiFID activity-related creditors do not have a greater entitlement to receive payments from the FOR capital than non-MiFID activity-related creditors, except where specifically provided for such subordination.

Linking the FOR to specific activities could in our view complicate the FOR-calculation process without providing substantial benefits. It could create unnecessary administrative burdens for firms and lead to inconsistencies in how firms prepare for potential wind-downs. Additionally, differentiating activities might not effectively address the practical realities of insolvency proceedings where all creditors, regardless of the type of activity, must be treated equally.

Therefore, we recommend maintaining a unified approach to calculating the FOR, without distinguishing between costs related to MiFID and non-MiFID activities. This approach ensures simplicity, clarity, and fairness in managing the capital requirements for the orderly wind-down of investment firms.

[1]            Which also seems to be in line with the original thinking of the IFR, see EBA Discussion Paper, (2016), p. 51.

Q8: Should expenses related to fluctuation of exchange rates be included in the list of deductions for the calculation of the FOR? If yes, which criteria should be considered in addition to the ones suggested above?

We do not have any comments on this question.

Q9: Should the concept of ‘ongoing advice’ be further specified for the purpose of calculating the K-AUM? If yes, which elements should be taken into account in distinguishing a recurring provision of investment advice from a one-off or non-recurring one?

The Discussion Paper raises the question whether the concept of “ongoing advice” needs further specification for the purpose of calculating K-AUM, and if so, what elements should be considered in distinguishing between recurring and non-recurring advice. In this context, we note that a recent Q&A of the EBA clarified when monitoring activities qualify as ongoing advice (and thus should be included in AUM), as certain technical services currently offered to investors are not straightforwardly identified as such.[1] The Q&A stipulates that:

1. assets related to monitoring activities that involve the mere reception and aggregation of statements on the value of assets under management by other undertakings should not be included in the AUM calculation;
2. assets related to monitoring activities that entail advice on the portfolios, such as the analysis of the performance of an investment portfolio (e.g., comparison with benchmarks and alternative scenarios analysis), asset allocation or concierge services, should be included in the AUM calculation.

Despite this, the demarcation of “monitoring services” remains somewhat unclear in our view. For instance, it is uncertain whether (assets related to) every monitoring service that goes beyond mere reception and aggregation of statements should automatically be included in AUM or if there needs to be an explicit element of advice.

To address these uncertainties, we propose a more precise and qualitative definition of “ongoing advice” that includes the requirement for a recurring, continuous, or non-incidental obligation to provide recommendations under a contract. This definition should also explicitly go beyond mere data aggregation and require an element of implicit or explicit suggestions on portfolio or investment decisions. This aligns with MiFID II’s distinction between providing mere information and providing recommendations, ensuring that only those assets covered by some sort of advisory relationship with a continuous element are included in AUM.[2]

Moreover, the Q&A suggests that if one firm provides monitoring activities and another provides advisory services, both firms should include the relevant assets in AUM. It is unclear to us whether the EBA, when referring to “monitoring activities” in this context refers to the monitoring activities under (i) above (i.e., those that involve the mere reception and aggregation of statements on the value of assets under managements) or under (ii) (i.e., those that entail advice on the portfolios). From the Q&A it appears to be the first, as the EBA refers to the potential risk of a misrepresentation or an inaccurate aggregation of the portfolios when referring to the risk of monitoring client’s portfolios. This would contradict the Q&A which stipulates that assets related to such monitoring activities should not be included in the AUM calculation. 

That being said, the Q&A seems to suggest double capitalisation for the same assets, which appears inconsistent with the original intent of AUM, which was determined as: ‘[…] for the firm that holds the responsibility to the customer to hold the adequate level of capital related to this activity under either AIFMD/UCITS or the new prudential regime.’

We propose the following:

1. If a firm performs monitoring activities that involve the mere reception and aggregation of statements on the value of assets under management, those assets should not be included in the AUM calculation, regardless of whether another firm provides advisory services in respect of the same assets.
2. Only the firm providing “ongoing advice” – which is directly responsible for managing the client’s investment decisions – should include the assets in AUM. This is based on the premise that the advising firm bears the primary responsibility for the client’s investment outcomes.

Thus, to reflect the original objective of K-AUM and the risk to the client it aims to cover, we recommend that, where ongoing investment advice and monitoring activities are performed by different firms for the same assets of a single client, only the firm providing the ongoing investment advice should include those assets in the AUM. 

[1]            EBA Q&A 2022_6418.

[2]            ESMA, Supervisory briefing on understanding the definition of advice under MiFID II, (2022), p. 10.

Q10: Does the K-DTF provide a proper level of capital requirements for the provision of the services Trading on own account and execution of order on behalf of clients on account of the investment firm? If not, what elements of the calculation of the K-DTF present most challenges?

1. Remain the current formula

Although we recognise the issues with the current K-DTF, we also believe that it is “the best of the rest”. The current K-DTF is a simple and proportional mechanism that provides for a capital buffer broadly related to the risks it targets. As such, we do not oppose retaining the current K-DTF.

    2. Change the notional amount to the market value

The current K-DTF framework includes the notional amount of derivative transactions, but this measure may not accurately reflect the operational risk associated with these trades.[1] The notional amount primarily seems to serve as a supervisory tool to gauge the scale of derivative contracts cleared by firms, particularly those that frequently net out or close their positions within a trading day. However, it does in our view not effectively capture the actual financial exposure or the operational risks involved in these transactions. Thus, if the current K-DTF calculation methodology were to be maintained, we recommend using a market value-based metric.

We propose amending art. 33(2)(b) IFR to replace the “notional value” of derivative contracts with the “market value” or the “amount paid or received” for each trade. 

We also suggest merging the separate coefficients for cash trades and derivative trades into a single (tiered) coefficient. This simplification will streamline the calculation and ensure consistency across different types of trades.

    3. Amend the scope of cash trades

To ensure that the K-DTF framework provides a consistent and accurate measure of operational risk, we propose clarifying the scope of transactions included under K-DTF. Currently, the K-Factors RTS specify that for cash trades, DTF should encompass transactions where a counterparty “undertakes to trade” certain instruments.[2] This wording has, in our experience, led to varying interpretations among firms, creating inconsistencies in how transactions are considered from a prudential perspective. Some firms include both settled and unsettled transactions, while others only include settled transactions. This lack of uniformity creates an unlevel playing field and introduces discrepancies in the application of capital requirements.

We propose clarifying which transactions are to be covered by the phrase “undertakes to trade” from art. 10(1) K-Factors RTS. 

[1]            As calculated pursuant to art. 29(3) IFR.

[2]            Art. 10(1)K-factors RTS.

Q11: Would you have any examples where the calculation of the K-DTF based on comparable activities or portfolios results in very different or counterintuitive outcomes? If yes, how could the calculation of the K-DTF be improved?

We consider the industry to be best situated to provide a clear example from practice.

Q12: What are the elements of the current methodology for the calculation of the K-ASA that raise most concerns? Taking into account the need to avoid complexifying excessively the methodology, how could the calculation of the K-ASA be improved to assess those elements?

The K-Factor for client assets safeguarded and administered (“K-ASA”) is designed to cover the risks associated with failures in the safeguarding process.[1] K-ASA ensures that firms hold capital proportional to the balances of client assets they hold, regardless of whether these assets are on the firm’s own balance sheet or in third‐party accounts.[2] ASA refers to the value of assets that an investment firm safeguards and administers for clients, regardless of whether these assets appear on the firm’s own balance sheet or are in third‐party accounts.[3]

However, this definition, along with art. 19(2) IFR, has created a problematic situation. Art. 19(2) IFR states: ‘Where an investment firm has formally delegated the tasks of safeguarding and administration of assets to another financial entity, or where another financial entity has formally delegated such tasks to the investment firm, those assets shall be included in the total amount of ASA which is measured in accordance with paragraph 1.’

Firstly, the terms “financial entity” and “formally delegated” are not defined in art. 4 IFR, leading to a certain degree of legal uncertainty regarding the scope of K-ASA.[4] Moreover, the scoping, where clear, seems to encompass circumstances that may not directly give rise to the risks K-ASA intends to cover. However, this cannot be directly deduced from the IFR itself, as the specific risks targeted remain rather vague. On the one hand, operational risks surrounding the safeguarding of client assets seems to be relevant, while on the other, liability risks arising from failures in safeguarding client assets also seem to play a role.

During the preparation of the IFR, the EBA explicitly referred to the organisational requirements of MiFID II related to safeguarding as the risks to be addressed by a prudential buffer through K-ASA.[5] Currently, the IFR does not differentiate between various types of safeguarding, such as structural or statutory segregation, nor the associated different levels of operational risk. Instead, the IFR treats assets safeguarded at a custodian bank the same as those retained on the firm’s own balance sheet, without considering the different operational tasks required in these scenarios.

As a result, in situations where MiFID II organisational requirements are met in a way that removes most safeguarding risks from the firm, K-ASA still imposes a capital requirement. This could be considered as a duplication of capital requirements, as in the case mentioned above, the custodian bank will have already capitalized the deposits according to the applicable prudential framework.

Therefore, it appears that K-ASA, in its current form, oversimplifies the organisational governance applied to safeguarding by firms in safeguarding assets, resulting in a severely risk-insensitive capital requirement. Additionally, when applied strictly, K-ASA might pose problems for small firms wishing to qualify as a Class 3 firm, as they will quickly have a non-zero K-ASA.[6]

Q12: What are the elements of the current methodology for the calculation of the K-ASA that raise most concerns? Taking into account the need to avoid complexifying excessively the methodology, how could the calculation of the K-ASA be improved to assess those elements?

In view of the above, we recommend differentiating between structural and statutory separation. Structural separation involves using a separate legal entity for safeguarding customer funds, whereas statutory separation involves a legal structure to legally separate client assets from the firm’s own assets. The latter scenario poses a higher risk of comingling client assets than the former, as structural separation requires strict administrative segregation (and not also separate legal segregation since the assets are already legally separated by the mere fact that they are held by a separate legal entity). Additionally, the separate legal entity is often a bank or another entity not under the firm’s control, which generally applies its own ‘additional’ separation procedures, which decreases the likelihood of operational risks materializing.[7]

In contrast, statutory separation requires both rigorous administrative segregation and legal segregation, imposing an additional operational burden on the investment firm with potentially significant consequences if not implemented correctly. Hence, different coefficients should be introduced in art. 15(2) IFR, as has been done for client funds through the different K-CMH coefficient for funds held in separated accounts and non-separated accounts.

[1]            Art. 14 IFR; EBA Discussion Paper, (2016), par. 37(c).

[2]            Recital (24) IFR.

[3]            Art. 4(1)(29) IFR.

[4]            B.J. Nieuwenhuijzen, (2021), par. 414.

[5]            Annex to the Opinion, (2017), par. 141.

[6]            B.J. Nieuwenhuijzen, (2021), par. 413.

[7]            However, where the firm uses a dependent legal entity, such double application of administration requirements does not really provide any additional protection for the customer, as it will be the firm itself that is responsible for the separation at the dependent legal entity; see B.J. Nieuwenhuijzen, (2021), par. 167-171.

Q13: Clients’ asset protection may be implemented differently in different Member States. Should this aspect be considered in the calculation of the K-ASA? If so, how should that be taken into account in the calculation?

We refer to our differentiation between the statutory and the structural methods of asset segregation above. In our view, the workings of a national statutory separation mechanism could best be addressed through a Pillar 2 assessment. The national regimes are too diverse to be fully incorporated into the Level 1 text, so we propose using the two general terms of asset safeguarding.

Q14: Should crypto-assets be included into K-factor calculation, either as a new K-factor or as part of K-NPR?

No incorporation of crypto-asset exposures into the K-factor calculation

We believe that incorporating crypto-asset exposures into the K-factor calculation, such as K-NPR, would be challenging. Currently, the market generally understands that proprietary trading in crypto-assets does not require authorization under the Markets in Crypto-Assets Regulation (“MiCAR”).[1] Moreover, MiCAR does not contain any specific prudential capitalisation requirements for the (proprietary) trading of crypto-assets, even if it were to be subject to an authorisation requirement.[2]

The Basel standard on Crypto-Assets[3], as provisionally included in art. 501d CRR3, does not appear to apply to firms, as this article is not included in a part of the CRR applicable to firms dealing on own account through art. 21 IFR. Consequently, K-NPR should only cover crypto-asset exposures that qualify as financial instruments through the market risk framework of CRR. For instance, derivatives with crypto-assets as underlying do qualify as financial instruments, which prompts the question as to how crypto-assets should be dealt with in this frame. Arguably, the most appropriate treatment currently available, within the constraints of the market risk framework, would be to mark crypto assets as commodities, expressing their volatile nature. Nevertheless, the uncertainty remains a significant issue for firms wishing to deal in crypto assets, with no supervisory guidance available.

Services related to crypto assets should be prudentially regulated under the MiCAR framework. Including crypto-assets in the K-factor calculation could create level playing field issues. Crypto-asset service providers (“CASPs”) authorized under MiCAR are not subject to an equivalent K-factor regime and must only meet the higher of PMC or FOR. Apparently, the EU legislator did not consider it necessary to create a K-factor-like framework under MiCAR. Therefore, introducing a new K-factor for crypto-asset services is primarily not advisable on dogmatic grounds. f at all, a K-factor requirement for crypto-assets should be implemented into MiCAR.

Including a K-factor regime for trading in crypto assets under MiCAR

If crypto-assets were to be included into a K-factor calculation, we would recommend doing so within the MiCAR framework rather than the IFR/IFD framework. From a methodological standpoint, we oppose introducing additional K-factor requirements for crypto-asset under the IFR/IFD, as this framework is designed for investment services and activities related to financial instruments. The RtM K-factors seek to protect the traditional financial market, not the crypto-asset market. It is therefore in our view not appropriate to apply K-factors to other activities than investment services and activities, e.g., including crypto-asset services.

Should it be decided to include crypto-assets in the K-factor calculation, we propose (i) introducing a new K-factor for crypto-assets in the MiCAR framework, and (ii) including a provision in the IFR stipulating that such K-factor applies to firms having crypto-assets in their trading book. In our view, this should be the most methodological sound approach towards the prudential treatment of crypto assets for investment firms.

Regardless, as a last resort to combat the legal uncertainty that is currently rife, as well as to close the prudential gap that is faced by crypto asset exposures on firms’ balance sheet, a new K-factor could be included in IFR.

Including a K-factor regime for trading in crypto assets in IFR

Crypto-assets on firms’ balance sheets can carry significant market risk. Therefore, it may be imprudent to leave the trading of crypto-assets entirely uncapitalised, especially when conducted by non-bank entities.

Whereas the European legislator may not have deemed it necessary to subject CASPs engaged in proprietary trading to any additional capital requirements, this does not mean that firms should be exempt from such requirements. If one were to rank CASPs, firms and banks by their importance to financial stability, CASPs would rank lowest and banks highest, with firms falling somewhere in between. Consequently, it is conceivable that firms should be required to capitalise their crypto-asset exposures but CASPs should not. Similarly, banks should be held to the most conservative and prudent standards according to the Basel standard on crypto-asset exposures, while firms could be subject to a prudent, but slightly less strict capital requirement.

Therefore, a specific regime for crypto-assets could be included in IFR. As the Discussion Paper suggests, this could either be done in a new K-factor or K-NPR. We do not consider the latter appropriate because K-NPR refers to CRR, which already contains the Basel standard on crypto-assets. Integrating crypto-assets into K-NPR would require importing the entire market risk framework into IFR, potentially creating a parallel regime with minor differences. This could lead to interpretation challenges and significantly complicate the IFR.

Thus, creating a new K-factor for crypto-assets, e.g. ‘K-CRY’, seems the preferred course of action. This K-CRY could be based on the structure of the CRR market risk framework, for instance on that applied to commodities under the (simplified) standardised approach. Such framework would allow for the netting and hedging of crypto-asset exposures, though the exact calibration of the coefficients applied in course of the K-CRY regime would be challenging. These coefficients should adequately reflect the volatile and risky nature of crypto-assets, however, they should not be inhibitive for the proprietary trading of crypto-assets. The level of these coefficients should be set by means of a quantitative survey, taking into account the different risk-profile of firms and banks and the characteristics of the crypto-asset markets.

If K-CRY is introduced, to maintain a level playing field with proprietary traders in crypto-assets (which is, arguably, currently unregulated, and, thus, not a capitalized activity), an inevitable consequence of applying this K-factor to MiFID proprietary traders is that dealing on own account in crypto-assets should be regulated under MiCAR and possibly subject to the same K-CRY requirements. Whether this is a desired consequence is beyond the scope of this response.

[1]            Regulation (EU) 2023/1114 of the European Parliament and of the Council of 31 May 2023 on markets in crypto assets.

[2]            See J.J.F. van der Meer and S.W. van de Ven, De implicaties van MiCAR op handelaren voor eigen rekening in cryptoactiva, (2024); and A. Pantaleo and A. Vianelli, Digital Asset Market Making and MiCAR: In or Out? (2024); though it appears that ESMA has not definitively made up its mind on this point at the time of writing.

[3]            BCBS, Prudential treatment of cryptoasset exposures, (2022), (the “Cryptoasset Standard”); and as amended by BCBS, Cryptoasset standard amendments, (2024).

Q15: In the context of addressing operational risk for investment firm trading on own account, is there any further element to be considered to ensure that the requirements are proportionate to their trading activities?

We refer to our answer under Question 10.

Q16: The discussion paper envisages the possibility to rely on alternative methodologies with respect to the K-DTF. If the respondents suggest an alternative approach, how would this refer to the two activities addressed under the K-DTF (trading on own account and execution on own account on behalf of the clients)?

To calculate the operational risk requirement, the Discussion Paper suggests (i) reverting to the Basic Indicator Approach (“BIA”), (ii) the Standardised Approach (“OP-SA”) of the CRR, or (iii) an option where firms add their FOR to the K-factor requirement as a capital charge for operational risk.

First, we note that the operational risk framework of the CRR will be fully revised with the implementation of CRR3, where the current approach for calculating operational risk will be replaced with the Business Indicator Component (“BIC”). Second, the current operational risk calculation methodologies, as the Discussion Paper identifies, cover too broad a scope to properly reflect the operational risks related to the specific activity of dealing on own account. Therefore, we consider remaining the K-DTF requirement discussed above, in combination with a firm specific Pillar 2 add-on, insofar necessary, more appropriate than (a simplification of) the CRR3 operational risk framework.

Q17: When addressing other activities an investment firm may perform, which elements, on top of the discussed ones, should be also taken in consideration?

We oppose the introduction of a new K-Factor to address other activities an investment firm may perform (e.g., crowdfunding service) and recommend that the risk of such activities is capitalized through the firm-specific assessment under Pillar 2, insofar the risks are not already capitalized through Pillar 1.

Q18: Investment firms performing MiFID activities 3 and 6 (trading on own account and underwriting on a firm commitment basis) are more exposed to unexpected liquidity needs because of market volatility. What would be the best way to measure and include liquidity needs arising from these activities as a liquidity requirement?

Currently, liquidity requirements for firms are specified in art. 43 IFR, which mandates that firms retain one-third of their FOR in liquid assets. Liquid assets are defined in art. 43(1)(a)-(d) IFR and generally align with HQLA under the CRR framework, along with unencumbered short-term deposits.[1] Additionally, firms may face further liquidity requirements under Pillar 2 to address their specific liquidity risk, as detailed in Commission Delegated Regulation (EU) 2023/1651 (“Liquidity RTS”).[2]

Overall, the liquidity requirement framework under IFR is considerably simpler compared to the Liquidity Capital Requirement (“LCR”) banks must follow, both in qualitative and quantitative terms. The primary liquidity need for the average firm is expected to be for regular operational expenses.[3] Unlike banks, firms typically do not face severe liquidity outflows during stress events, such as bank runs.

To address the lower liquidity requirement, the same haircuts applicable to banks are applied to liquid assets under the IFR, providing a level of inherent prudence. Despite its simplicity and comparatively lower requirement, the liquidity framework appears appropriate for most firms. Unlike, the LCR[4], which is a stress-based metric, the IFR liquidity requirement focuses on the fundamental principle that investment firms are non-systemic and can be orderly wound down if necessary.[5]

The Discussion Paper explores several aspects of calibrating the IFR liquidity requirement, generally advocating for increased liquidity and additional qualitative clarifications.

Q18: Investment firms performing MiFID activities 3 and 6 (trading on own account and underwriting on a firm commitment basis) are more exposed to unexpected liquidity needs because of market volatility. What would be the best way to measure and include liquidity needs arising from these activities as a liquidity requirement?

In drafting the IFR, specific attention was given to the liquidity needs of firms dealing on own account and underwriting on a firm commitment basis. Historically, under the CRR, these firms were the only investment firms subject to liquidity requirements.[6] Their liquidity is particularly sensitive to market risk, as sudden price movements may necessitate additional margin deposits.[7] The EBA determined that a differentiated liquidity regime for firms dealing on own account should be addressed through Pillar 2, while applying a uniform Pillar 1 liquidity requirement to all firms.[8]

The most (conservative) expression of the liquidity requirement for most firms dealing on own account is the minimum net liquidity they must retain at their clearing members. This amount typically includes a safety margin beyond the market risk charge determined by the clearing member and should cover most trading-related liquidity needs. For operational expenses, firms should maintain cash reserves at their bank, ideally with the same institution.

The margin retained at the clearing member serves as a stress-based buffer, reflecting the firm’s specific liquidity needs. Conversely, cash held in a cash account covers operational expenses and is less likely to fluctuate significantly under stress.

Trading capital is often funded by equity, so depositing this capital as cash at the clearing member effectively creates a liquidity buffer. In that sense, firms already ring-fence a part of their capital through depositing it at the clearing bank. However, clearing and cash accounts should not be part of the same netting set to avoid encumbrance issues. This separation ensures that liquidity used for operational expenses remains unaffected by margin calls. This could be construed as an encumbrance of the deposit, as clearing banks generally may use such deposits to remediate any capital shortages in the clearing account. In other words, when the firm would need the liquidity to cover an operational expense, but the clearing member requires additional margin, generally speaking the clearing member may debit the cash account held to the required amount. Therefore, the liquidity requirement as currently applicable should be kept on a completely separate bank account, or if held in liquid assets, in a separate securities account. As such, the liquidity held to adhere to the liquidity requirement is ‘dead’ capital for a trading firm, it cannot add to the liquidity available for trading. 

Maintaining one month of operational expenses as a liquidity buffer is standard and essential for business continuity. However, requiring several months’ worth of liquidity could be economically burdensome and impractical, as it would need to be funded through equity, bonds or loans. Moreover, firms are generally prohibited from using the liquidity to meet the requirement, unless exceptional circumstances occur and supervisory approval has been obtained.[9] This immobile nature of the liquidity required was one of the reasons in the preparation of the IFR to moderate the quantity of liquidity required, so as to not crowd well-managed firms out of the market through regulatory reasons.[10]

A tailored regime for firms dealing on own account could be based on their trading activities and liquidity profile, potentially using a volatility percentage with a buffer over a one year period in respect of the RtM K-factor requirement. However, this approach might better reflect market risk than liquidity risk and would be a poor substitute for the clearing member’s margin model. In the end, the volatility in margin calls may be a better metric but this is highly unpredictable and firm-specific, and firms do not (necessarily) report margin calls to competent authorities, complicating this approach. The EMIR Review may improve transparency on margin calls under stressed conditions, providing insight into appropriate liquidity buffers. If a margin based buffer were implemented, it could result in a double requirement, as liquid assets should not be part of the clearing account’s netting set. Moreover, such margin based requirement would not acknowledge the liquidity providing character of most if not all clearing relationships, with the clearing bank generally providing for a liquidity facility for the trading firm.

Given the complexities and firm-specific nature of margin-based metrics, it is preferable to rely on the SREP for liquidity adequacy assessments. The SREP can tailor liquidity requirements based on the firm’s trading strategy, a more flexible and precise approach than a fixed Pillar 1 requirement.[11] Admittedly, this reasoning is severely focused on cleared trading; however, OTC trading lends itself even less to a predetermined Pillar 1 regime and is less predictable than cleared markets.

Raising the liquidity requirement to several months of fixed costs, instead of one, would further immobilize capital that could be deposited as margin, due to the prohibitive function of art. 44(1) IFR. This would strain firm funding and is an inappropriate proxy for the liquidity needs due to trading stress. Thus, we agree with the EBA’s conclusions in 2017 and find developing a specific liquidity regime for firms dealing on own account to be impractical and undesirable.

[1]            Whereas banks may not use commercial bank deposits as liquid assets and are restricted to certain central bank deposits, investment firms are permitted to use such liquidity to meet their requirements. This as for investment firms there is not a similar contagion risk or interconnectedness problematic as cross holdings under banks pose; see Annex to the Opinion, (2017), par. 252-253; EBA Discussion Paper, (2016), par. 117.

[2]            Commission Delegated Regulation (EU) 2023/1651 of 17 May 2023 supplementing Directive (EU) 2019/2034 of the European Parliament and of the Council with regard to regulatory technical standards for the specific liquidity measurement of investment firms under Article 42(6) of that Directive.

[3]            Annex to the Opinion, (2017), par. 231.

[4]            Which was one of the crucial points being considered in the drafting of the IFR; See EBA, Report on Investment Firms, (2015), p. 48; EBA Discussion Paper, (2016), par. 127-129; Annex to the Opinion, (2017), par. 236-141.

[5]            Commission Staff Working Document, (2017), p.9: ‘As a result of the low systemic relevance of most investment firms, the underlying CRR/CRDIV premise of ensuring sufficient buffers of capital and liquidity to withstand significant losses may be misplaced and represent disproportionate costs for them.’; Recital (28) IFR; and Discussion Paper, (2024), par. 177.

[6]            EBA, Report on Investment Firms, (2015), p. 44.

[7]            EBA, Report on Investment Firms, (2015), p. 46; Annex to the Opinion, (2017), p. 72.

[8]            Notwithstanding the possible derogation for class 3 firms under art. 43(3) IFR.

[9]            Art. 44 IFR; the Annex to the Opinion, (2017), par. 251 considered it best not to subject the use of the liquidity buffer in exceptional circumstances to prior supervisory approval, however, the IFR does seem to have implemented such a prior approval regime. This makes it even harder for a firm to use the liquidity and in the event of a sudden margin call, which is discussed here, the firm will have to obtain supervisory approval before the final margin deposit time – generally no later than end of day. This makes art. 44(1) IFR particularly troublesome in the event of liquidity stress. 

[10]          EBA Discussion Paper, (2016), par. 128.

[11]          EBA Discussion Paper, (2016), par. 129; See also art. 2(2) of Commission Delegated Regulation (EU) 2023/1651.

Q19: Investment firms performing the activities of providing loans and credit to clients as an ancillary service in a non-negligeable scale would be more exposed to liquidity risks. What would be the best way to measure such risk in order to take them into account for the purposes of the liquidity requirements?

Providing loans and credit to clients is not a common activity for most firms and generally does not occur on a large scale.[1] However, some firms may engage in this activity more significantly, though they represent a small subset of the total investment firm population. As mentioned earlier regarding the Pillar 1 capital charge for these activities, introducing a Pillar 1 liquidity requirement would complicate the IFR to the extent that it would undermine the proportionality principle that underpins it.

We would advise against the introduction of a flow or stock-based measure, as this would effectively create a quasi-LCR, shifting the Pillar 1 regime closer to the banking regime – something that has been deemed inappropriate and disproportionate. Introducing a provision similar to art. 45 IFR, which addresses client guarantees, would be challenging. Client loans are on-balance-sheet items and imposing a provision akin to art. 45 IFR would essentially introduce a credit risk weight for these client exposures in the form of liquidity rather than capital. Moreover, this approach would disrupt the IFR framework by requiring specific liquidity requirements for certain on-balance-sheet items while not doing so for others.

Therefore, we believe it to be the most appropriate to continue handling the treatment of client loans under the SREP, given the highly idiosyncratic nature of such lending. Additionally, under the qualitative requirements of art. 29(1)(d) IFD, there is a sufficiently prudent risk management framework in place to manage the liquidity risks that an investment firm may face.

[1]            Commission Staff Working Document, (2017), p. 16.

Q20: Investment firms, providing any of the MiFID services, but exposed to substantial exchange foreign exchange risk may be exposed to liquidity risks. What would be the best way to measure such risk in order to take them into account for the purposes of the liquidity requirements?

We believe that the best way to measure the foreign exchange risk (“FX risk”) to which firms are exposed would be (a proportional version of) the method as used by art. 351 et seq. CRR. 

Technically, art. 21(4) IFR already requires the inclusion of positions, beyond those in the trading book, which give rise to FX or commodity risk in their own funds requirements through the RtM K-factor requirement. However, it is unclear to us whether firms that do not deal on own account are required to capitalize their FX risk under this provision. Applying this requirement to all firms could also be disproportionate and unfamiliar for many firms that do not deal on own account, as this means that all firms would need to adhere to the FX risk framework outlined in art. 351 et seq. CRR. These firms, however, would be the primary targets for an FX risk liquidity charge, as those dealing on own account already quantify and capitalize their FX risk through their operational funding, utilizing clearing accounts and margin requirements.

One potential solution is to implement a proportional liquidity requirement version of the FX risk capital requirement from the CRR into the IFR. The detailed and complex rules of the LCR, including those on currency mismatches, are too intricate to apply effectively to all firms. Therefore, adapting the capital requirement under art. 351 CRR into a proportional liquidity measure could be beneficial, as this methodology is relatively straightforward. It is important to note that qualitative requirements regarding FX liquidity risk, such as those found in art. 29(1)(d) and, to some extent, art. 42 IFD, are essential complements to any Pillar 1 requirement.

To address this, a new article could be included:

‘Article 43a

Foreign exchange liquidity requirement

1. If the sum of an investment firm’s overall net foreign-exchange position, calculated in accordance with paragraph 3, exceeds 10% of its total own funds, the investment firm shall calculate, in addition to the requirement of Article 43(1), a liquidity requirement for foreign exchange risk. The liquidity requirement for foreign exchange risk shall be the sum of its overall net foreign-exchange position in the reporting currency, calculated in accordance with paragraph 3, multiplied by 4%.

2. The net foreign exchange position in each currency other than the reporting currency shall be calculated as the sum (positive or negative) of asset items (positive) and liability items (negative). 

3. The net position in foreign exchange per currency other than the reporting currency, calculated in accordance with paragraph 1, shall be converted at spot rates into the reporting currency. They shall then be summed separately to form the total of the net short positions and the total of the net long positions respectively. The higher of these two totals shall be the investment firm’s overall net foreign-exchange position.

4. The requirement calculated in accordance with paragraph 1 shall be held in liquid assets as set out in points (a) to (d) of Article 43(1).

5. By way of derogation from paragraph 1, competent authorities may exempt investment firms that meet the conditions for qualifying as small and non‐interconnected investment firms set out in Article 12(1) from the application of paragraph 1 and shall duly inform EBA thereof.’

Q21: Are there scenarios where the dependency on service providers, especially in third countries, if disrupted, may lead to unexpected liquidity needs? What type of services such providers perform?

We do not have any comments on this question.

Q22: Are there scenarios where the dependency on liquidity providers, especially in third countries, would lead to unexpected liquidity needs? Could you provide some examples?

We do not have any comments on this question.

Q23: What other elements should be considered in removing the possibility of the exemption in Article 43 of the IFR?

For the reasons outlined above and mentioned in the Discussion Paper, we support removing the exemption for small and non-interconnected firms currently included in art. 43(1), second sub-paragraph, IFR. Small firms, like larger ones, need to cover their monthly operational expenses and should be capable of being wound down in an orderly manner. Therefore, whether one adopts a going concern or gone concern approach to liquidity requirements, there does not appear to be a compelling reason to exempt small and non-interconnected firms. However, these firms should still be allowed to use the assets listed in art. 43(3) IFR, considering the specific nature of their business models.[1]

We would also like to address art. 44(1) IFR. This provision deviates from the original proposals by the EBA and the Commission.[2] Initially, the provision was intended to impose a general prohibition on using the liquidity buffer except in exceptional circumstances. If a firm invoked this provision, it was required to notify the competent authority immediately and restore the full requirement within 30 days.[3] However, in the final and current IFR, the notification requirement was changed to a prior approval right. Reflecting on the EBA’s rationale, this design choice seems illogical to us: “Investment firms should be allowed to use the liquidity buffer only in exceptional and unexpected circumstances. As liquidity may be required in a very short term, the use of the liquidity resources may be allowed prior to supervisory approval; however, this should always be immediately accompanied by notification to the competent authority.”[4]

This change appears to have been made after the European Parliament’s report and was likely agreed upon during trilogue negotiations, as it is included in the final text.[5] We recommend removing this amendment, as the liquidity buffer should be accessible to firms when needed. Particularly in the case of unexpected liquidity shocks, which are typically what drive the need for a buffer, firms may need to act within a single day, such as for margin calls. Waiting for supervisory approval in such cases seems to create an unnecessary obstacle without adding significant prudential value.

Therefore, we propose redrafting art. 44(1) IFR as follows:

‘1. Investment firms may, in exceptional circumstances, reduce the amount of liquid assets held. Where such reduction occurs, the investment firm shall notify the competent authority without delay.’

[1]            Annex to the Opinion, (2017), par. 255-256. 

[2]            See art. 43(1) IFR Proposal; and Commission Staff Working Document, (2017), p. 22.

[3]            Art. 44(2) IFR.

[4]            Annex to the Opinion, (2017), par. 250.

[5]            Art. 43(1) of European Parliament, Provisional Agreement Resulting from Interinstitutional Negotiations: 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)0790 – C8-0453/2017 – 2017/0359(COD)), (2019).

Q24: Do you have any views on the possible ways forward discussed above concerning the provision of MiFID ancillary services by UCITS management companies and AIFMs?

Background

The Discussion Paper states that it is essential to identify regulatory loopholes that allow entities to conduct investment firm activities or provide investment firm services without being covered by the IFR/IFD. In this context, it refers to art. 6(3) of Directive 2009/65/EC (“UCITS Directive”) and art. 6(4) of Directive 2011/61/EU (“AIFMD”), pursuant to which a UCITS management company or AIFM (“Fund Manager”) may provide the following MiFID services:

1. Individual portfolio management;
2. Non-core services comprising:
   1. Investment advice;
   2. Safe-keeping and administration in relation to shares or units of collective investment undertakings.

Additionally, AIFMs may also provide reception and transmission of orders in relation to financial instruments (together, with the services under (a) and (b): “MiFID Top-Up Services”). When a Fund Manager is authorized to provide MiFID Top-Up Services, it must comply with the relevant MiFID II conduct rules that apply to such services. However, under MiFID II and the IFR, no capital requirements arise in relation to the provision of these additional services.[1] As a result, Fund Managers can be authorised, on top of their UCITS/AIFMD-specific services, to carry out MiFID services, without their own funds requirements under the UCITS Directive and AIFMD necessarily taking due account of the risks associated with these services. As also noticed by the Discussion Paper, this may result, in certain cases, in an asymmetric prudential treatment of these services compared to investment firms providing the same services, e.g., by having to set aside regulatory capital under relevant K-factors.

Recommendations

To ensure a symmetric and adequate level-playing field, we recommend that Fund Managers authorized to provide MiFID Top-Up Services should comply with the same prudential rules as investment firms that provide the same services. The reason therefore is that providing investment services involves certain prudential risks, regardless of whether the entity providing the services is authorized under a MiFID licence, an AIFMD licence or a UCITS licence. It is therefore evident that the approach to addressing these risks is independent of the specific licence under which the services are provided. Aligning prudential rules applicable to Fund Managers for the provision of investment services with the prudential rules applicable to investment firms, ensures that Fund Managers providing investment services maintain sufficient capital for the risks associated with the provision of these services (“same risks, same rules”). Additionally, this prevents regulatory arbitrage: entities cannot influence the prudential rules applicable to the investment services they provide by selecting a specific licence.

More concretely, we recommend a prudential twofold approach[2], whereby Fund Managers authorized to provide MiFID Top-Up Services are subject to the highest of the capital requirements (i) as calculated under the AIFMD/UCITS-regime, and (ii) as calculated under the IFR-regime. This approach prevents overlap and double counting between the prudential requirements of the AIFMD or UCITS Directive and the IFR, where the same risks might be mitigated by two capital requirements. In practice, this increases the regulatory burden for Fund Managers providing MiFID Top-Up Services. However, since only the highest capital requirement would apply, it is crucial that (i) capital requirements under the IFR and the AIFMD or UCITS Directive are fully calculated by the Fund Managers, and (ii) determining the capital requirement under the IFR is done independently from the determination under the AIFMD or UCITS Directive and vice versa. Potential overlap between IFR capital requirements and those under the AIFMD or UCITS Directive is also avoided by applying only the highest capital requirement.

Practical implications

In practice, under the AIFMD/UCITS-regime, a Fund Manager authorized to provide MiFID Top-Up Services should maintain at least the highest of the following capital requirements:

1. a minimum capital requirement for Fund Managers of EUR 125,000[3];
2. a FOR of 25% of the fixed costs over the past financial year[4];
3. an AUM requirement of EUR 125,000 plus 0.02% times the AUM above EUR 250 million[5].

Under the IFR-regime, Fund Managers should then maintain at least the highest of the following capital requirements:

1. the permanent minimum capital requirement for a firm (i.e., Fund Manager) that performs MiFID Top-Up Services of EUR 75,000[6];
2. a FOR of 25% of the fixed costs over the past financial year[7];
3. the K-factor requirement, being the applicable requirement as an exponent of the relevant K-Factors of the Fund Manager[8].

Consequently, the higher of the AIFMD/UCITS-regime and the IFR-regime would be the leading quantitative capital requirement for the Fund Managers providing investment services. This method is in our view also intuitive from a methodological point of view, in the sense that a Fund Manager that provides relatively limited MiFID Top-Up Services (compared to its AIFMD/UCITS business) will have relatively low K-factors and therefore more likely fall under the AIFMD-regime. However, a Fund Manager that provides a more significant amount of MiFID Top-Up Services (compared to its AIFMD/UCITS business) will have relatively high K-factors and therefore more likely fall under the IFR-regime.

Class 3 Fund Managers

In view the rationale behind Class 3 firms, we believe that for Fund Managers that qualify as Class 3 firms under art. 12 IFR, the IFR-regime capital requirement should be the highest of the FOR and the permanent capital requirement. As a result, a Fund Manager that meets the requirements under art. 12 IFR should not need to calculate the K-factor requirement.

K-factors

The K-factor requirement would be the main difference between the AIFMD/UCITS regime and the IFR regime. Fund Managers should in our view only calculate K-factors for services attributable to the investment services they provide. In view of the investment services that Fund Managers may provide under the AIFMD and the UCITS Directive, the RtM and RtF K-factors do not apply to Fund Managers. Consequently, the only relevant K-factors are K-AUM, K-COH, K-CMH and K-ASA.

K-AUM

For Fund Managers that provide MiFID Top-Up Services, K-AUM likely is the most relevant K-Factor. As set out under paragraph 3.2,-AUM pertains to the value of assets that a firm manages for its clients under both discretionary and non-discretionary portfolio management arrangements constituting investment advice of an ongoing nature. Discretionary investment management refers to managing portfolios in accordance with mandates given by clients on a discretionary basis as defined in art. 4(1)(8) MiFID II, which means managing portfolios based on client instructions containing one or more financial instruments. Therefore, “collective” portfolio management (as in scope of the AIFMD and UCITS Directive) does not fall under discretionary portfolio management (as in scope of the MiFID/IFR-regime).

Under the AIFMD and UCITS Directive, a Fund Manager must maintain own funds of EUR 125,000 plus 0.02% of the amount exceeding EUR 250 million, up to a maximum of 10 million. We can imagine that AUM-related capital requirements under the IFR and those under the AIFMD and the UCITS Directive overlap, particularly if Fund Managers invest “individually” managed (i.e., AUM) in the “collective” funds they manage. It is therefore essential that the capital requirements under the IFR and the AIFMD or UCITS Directive are fully calculated and that no netting occurs with the capital requirement for collectively managed portfolios under the AIFMD or UCITS Directive when calculating the K-AUM requirement.

K-COH

As referred to in paragraph 3.1, COH concerns the value of orders handled for clients, through the reception and transmission of client orders and through the execution of orders on behalf of clients. Since a Fund Manager cannot provide the investment service of execution of orders on behalf of clients, the calculation of K-COH only applies to the reception and transmission of client orders. For Fund Managers, K-COH should in our view only pertain to orders from individual clients, meaning that orders handled by a Fund Manager in the context of managing a client’s investment portfolio should not be covered by K-COH if the Fund Manager has already calculated K-AUM for that client’s investments. K-COH should include transactions arising from investment advice, insofar the investment firm did not calculate K-AUM in respect of those transactions. Consequently, orders received and transmitted in connection with ongoing investment advice, for which the manager calculates K-AUM, should not be included in K-COH.

K-CMH

The K-factor K-CMH applies when a Fund Manager holds client money. If the Fund Manager is not authorized to conduct the MiFID ancillary service of safekeeping, we believe that no capital should be required based on K-CMH. However, if the Fund Manager has such authorization, e.g., for purposes of asset segregation requirements, the Fund Manager should calculate K-CMH.

K-ASA

The K-factor K-ASA applies when a Fund Manager safeguards and administers assets for clients from individual investment mandates. If the Fund Manager is not authorized for the MiFID ancillary service of safekeeping, we believe that no capital should be required based on K-ASA. However, if the Fund Manager has such authorization, e.g., for purposes of asset segregation requirements, the Fund Manager should calculate K-ASA.

[1]            Except for local law differences. For example, under Dutch law, Fund Managers are subject to IFR capital requirements in relation to the provision of MiFID services.

[2]            Similar to the twofold approach currently applicable to Fund Managers under Dutch law.

[3]            Art. 9(2) AIFMD / art. 7(1) UCITS Directive.

[4]            Art. 9(5) AIFMD / art. 7(1)(a)(iii) UCITS Directive.

[5]            Art. 9(3) AIFMD / art. 7(1)(a) UCITS Directive.

[6]            Art. 14 IFR jo. 9(2) IFD.

[7]            Art. 13 IFR.

[8]            Part Three IFR.

Q25: Are differences in the regulatory regimes between MICAR and IFR/IFD a concern to market participants regarding a level playing field between CASPs and Investment firms providing crypto-asset related services? In particular, are there concerns on the capital and liquidity requirement regimes?

The Discussion Paper addresses the interaction between the MiCAR and the IFR/IFD framework.[1] It examines how MiCAR interacts with the capital and liquidity requirements under IFR/IFD for firms providing crypto-asset services.[2] The Discussion Paper notes that firms may be exempt from most (if not all) prudential requirements for CASPs if they follow the notification regime under art. 60 MiCAR. As the Discussion Paper indicates, this exemption arises because the IFR/IFD regime is generally considered more conservative than MiCAR. The situation is in that sense comparable to banks providing investment services, which are not subject to IFR/IFD requirements because CRR/CRD IV requirements are generally more stringent.

The question then is how IFR/IFD prudential requirements apply when a firm acts as a CASP. Specifically, should crypto-asset services be included in the K-Factors, the FOR and/or the liquidity requirement?

We believe that, in principle, no additional Pillar 1 requirements should be applied to crypto-asset services under the IFR/IFD. First, the IFR/IFD framework is designed for investment services and activities. As previously mentioned, we do not consider it appropriate to use Pillar 1 to target other activities than investment services and activities. Therefore, crypto-asset services should be addressed within the MiCAR framework. Second, including CASP activities in Pillar 1, such as in the K-factors, could create level playing field issues. Pure CASPs are not subject to a K-factor equivalent regime and must only hold the higher of PMC or FOR.[3] Introducing a new K-factor for crypto-asset services is therefore not advisable; if at all, such requirement should be handled under MiCAR. That being said, if it is decided to introduce a K-factor for crypto-asset exposures, we propose to do so in the manner set out in our answer to Question 14.

Third, the FOR is based on a firm’s total expenditure. Excluding crypto-asset services from the FOR would require a new deduction ground. However, in line with our previous comments, the FOR should cover the orderly wind-down of the entire firm, not just its investment services and activities of the firm. Thus, excluding crypto-asset services from the FOR calculation under art. 13 IFR is not appropriate in our view. Fourth, we recommend a restrictive approach for the capitalisation of crypto-asset services within the Pillar 2 framework, where they would pose a risk that ought to be caught in the IFD based SREP assessment.[4] Class 1 and Class 1a firms must follow the transitional regime due to CRR/CRD IV requirements.

Competent authorities may identify risks related to crypto-asset services that need to be capitalised. However, MiCAR does not have a similar Pillar 2 regime for CASPs, leading to an uneven playing field if firms face a Pillar 2 add-on for their crypto-asset services. Moreover, MiCAR made a conscious legal choice not to impose additional capital requirements beyond the FOR or PMC. Nevertheless, the prudential requirements under art. 67 MiCAR aim to ensure consumer protection, similar to the RtC-induced capital requirements under the IFR.[5] Evidently, the EU legislator did not find it necessary to impose higher capital requirements for adequate consumer protection. The IFR, with its RtC K-Factor requirements specifically aimed at capitalising risks to clients, does necessitate specific capital for client protection. This discrepancy means that a MiCAR customer and an IFR/MiFID II client do not receive equal protection. Fundamentally, capitalising risks to clients for firms also providing crypto-asset services through Pillar 2 would differentiate in the protection offered to a service-recipient depending on whether they approach a CASP or a firm, creating an unlevel playing field. However, dogmatically, it would be difficult to permit firms to award a lower protection to clients when they receive crypto-asset services than when they receive investment services.

Despite these issues, we believe that Pillar 2 should not be used to capitalise all risks to clients from crypto-asset service provision. Although this creates the problems described above, the legislator seems to have opted for lower protection for crypto-asset service recipients. Pillar 2 add-ons should in our view only be applied if risks to customers adversely affect MiFID II clients, the traditional financial markets or the firm’s stability to the point of preventing an orderly wind-down.

[1]            Discussion Paper, (2024), par. 214-225.

[2]            Discussion Paper, (2024), par. 217-222; See art. 3(1)(16) MiCAR for a definition of crypto-asset services. Roughly, most of the crypto-asset services coincide with the non-token-based investment services and activities included in MiFID II, see art. 60(3) MiCAR.

[3]            Art. 67(1) MiCAR.

[4]            See Discussion Paper, (2024), par. 224.

[5]            Recital (80) MiCAR.

Q26: Sections 5.2, 5.4 as well as this Section 9.1 all touch upon how crypto-assets (exposures and services) may influence the IFD and the IFR. Is there any other related element that should be considered in the review of the investment firms’ prudential framework?

We do not have any comments on this question.

Q27: Is the different scope of application of remuneration requirements a concern for firms regarding the level playing field between different investment firms (class 1 minus and class 2), UCITS management companies and AIFMs, e.g., in terms of the application of the remuneration provisions, the ability to recruit and retain talent or with regard to the costs for the application of the requirements?

We do not have any comments on this question.

Q28: Are the different provisions on remuneration policies, related to governance requirements and the different approach to identify the staff to whom they apply a concern for firms regarding the level playing field between different investment firms (class 1 minus under CRD or class 2 under IFD), UCITS management companies and AIFMs, e.g. in terms of the application of the remuneration provisions, the ability to recruit and retain talent or with regard to the costs for the application of the requirements?

We do not have any comments on this question.

Q29: Are the different provisions, criteria and thresholds regarding the application of derogations to the provisions on variable remuneration, and that they apply to all investment firms equally without consideration of their specific business model, a concern to firms regarding the level playing field between different investment firms (class 1 minus under CRD and class 2 under IFD), UCITS management companies and AIFMs, e.g., in terms of the application of the remuneration provisions, the ability to recruit and retain talent or with regard to the costs for applying the deferral and pay out in instruments requirements? Please provide a reasoning for your position and if possible, quantify the impact on costs and numbers of identified staff to whom remuneration provisions regarding deferral and pay out in instruments need to be applied.

We do not have any comments on this question.

Q30: Are the different provisions regarding the oversight on remuneration policies, disclosure and transparency a concern for firms regarding the level playing field between different investment firm, UCITS management companies and AIFMs, e.g., with regard to the costs for the application of the requirements or the need to align these underlying provisions? Please provide a reasoning for your position.

We do not have any comments on this question.

Q31: What would be costs or benefits of extending existing reporting requirement to financial information? Which other elements should be considered before introducing such requirement?

We do not have any comments on this question.

Q32: Should there be the need to introduce prudential requirement for firms active in commodity markets and that are not currently subject to prudential requirements? How could the existing framework for investment firms be adapted for those cases? If a different prudential framework needs to be developed, what are the main elements that should be considered?

We do not have any comments on this question.

Attachment

Name of the organization