LGIM believes that the identification of ‘systemic and bank-like’ investment firms should be based upon the business models of firms; the most important prerequisite for such firms is a significant balance sheet exposure to other financial institutions. The vast majority of asset managers do not trade as principal or hold client assets, and so present a very limited level of systemic risk to the financial system. It is essential that such institutions fall outside scope of the criteria for ‘systemic and bank-like’ and are thus subject to a less complex regime, which is proportionate to the actual level of risk that they pose to customers and the financial system. The new prudential regime should therefore maintain a high threshold for classifying investment firms as ‘systemic and bank-like’, based upon the criteria under CRD IV.
Even the largest asset managers typically do not present a comparable level of systemic risk to the financial sector as do global banks, given that there is limited risk of contagion from the collapse of such firms. This means that government bail-outs are not required to prevent the public from harm when such businesses become unviable. Asset managers usually operate an agency business model and can therefore be wound up in an orderly manner without external adverse effects. This is because client assets are held by a third party depositary and so are placed completely out of reach of the firm’s creditors. Nevertheless, key to such safeguards working effectively is a well-designed prudential regime which ensures that client assets and monies are appropriately segregated, with asset managers capitalised to cover any loss caused by operational, market or other failures.
LGIM believes that application of the same criteria, as provided for G-SIIs and O-SIIs, for the identification of ‘systemic and bank-like’ investment firms would be appropriate provided that additional reporting requirements are not introduced for firms which are currently out-of-scope of FINREP reporting. Competent authorities should continue to be afforded the same discretion in identifying G-SII’s and O-SII’s as is currently in place under CRD IV.
Finally, LGIM views it as very important that investment firms which also provide custody services are regarded as systemic and bank-like. Given the significant balance sheet exposure that such institutions have in holding client assets, they would present a high risk to both customers and the wider market if they were to become insolvent. We recommend an explicit rule for the provision of custodian services to automatically satisfy the ‘client money held’ K-factor, and therefore bring the firm under Class 1.
LGIM recognises that the principles proposed in the EBA’s discussion paper are not fully developed and that a considerable amount of work still needs to be done before the new prudential regime can be applied across the industry. We agree with the concept of assessing capital requirements based upon the individual risks inherent in the business of investment firms, rather than adapting inappropriate measures which have been developed for large international banks. However, the detail in how these are defined and calibrated will be vital in ensuring that the capital requirements properly reflect the risks to customers, firms and the financial markets.
To complement this, we recommend that the new prudential regime also takes into consideration the differences in control environments operated by individual firms, as part of the ‘risk-to-firm’ assessment. Where a firm has robust controls in place which effectively manage the risks present in its business model, this should be captured in the regulatory assessment and reflected by a lower capital requirement. This may also promote good practices within the industry, prompting firms to bolster their control frameworks in order to take advantage of a less onerous application of prudential standards.
A proper definition of a firm’s risk appetite must also be incorporated into the regulatory assessment for the risk proxies to produce an appropriate capital requirement.
Finally, we envisage potential issues around the divergent nature of the definitions of “Own Funds” and “capital requirements”. Own Funds under CRD IV includes “audited reserves”. For investment firms which are subject to an annual audit only, this means that audited reserves comprising Own Funds are fixed once a year after the audit opinion has been concluded. However, under a K-factor approach which uses measurable data at a point in time, a firm’s capital requirements may increase intra-year as the underlying measurable components (such as assets under administration) increase. This means that investment firms would face increasing requirements without benefitting from the increase in net assets generated from growth in the business, as unaudited profits cannot be included in Own Funds under CRD IV. This could result in otherwise solvent firms having to raise additional capital.
In our view, this discrepancy could be addressed by allowing firms to include unaudited profits (with deduction of foreseeable dividends) within Own Funds; this would better reflect the loss absorbency capacity of investment firms at any point in time.
LGIM believes that differentiating small and non-interconnected investment firm as a separate class, subject to a fixed overhead requirement regime only, would be appropriate. This approach would result in achieving an appropriate level of prudential risk management within a relatively simple framework. We agree that, whilst a single prudential regime with built-in proportionality may produce a comparable overall capital requirement for such firms, the administrative burden of extracting and calculating a K-factor approach is likely to be disproportionate.
LGIM agrees that a combination of qualitative and quantitative factors should be taken into account in assessing which firms should be captured under a Class 3 classification.
Holding client money or securities:
The practice of firms holding client money varies widely across Europe, with some jurisdictions requiring client money to be held in segregated accounts where the firm has no legal or beneficial ownership. LGIM proposes that client money held in segregated accounts should not preclude firms from being captured under Class 3, as in such cases the investors’ monies are appropriately ring-fenced and protected in any insolvency scenario of the investment firm.
Dealing on own account:
LGIM believes that the MIFID activity of dealing on own account should preclude a firm from falling under Class 3, provided that positions taken where an asset manager operates as principal for operational order fulfilment or to hold capital/liquid resources are not included in the definition of dealing on own account.
The MiFID II activity of operating an organised trading facility (OTF):
LGIM’s view is that being member of a wider group should not preclude an investment firm from falling under Class 3 on a solo basis, provided that the firm is included in equivalent consolidated supervision at the overall Group level (such as Consolidated CRD IV, Solvency II or under the new investment regime).
LGIM agrees that these three categories broadly capture the risks that investment firms present in carrying out their business activities. However, concerning the risk to firm (RtF) category, we do not believe that the proprietary trading activities (PTA) consideration is relevant for the majority of firms. Asset managers falling into Classes 2 and 3 (i.e. those that are not systemic or bank-like) trade for the benefit of their clients rather than for their own gain. If PTA is taken forward as a K-factor, it should be expressly stated that this only applies to firms which undertake such activity, and therefore be limited to Class 1 firms.
In addition, LGIM believes that the RtF category could be used to allow either the firm itself and/or competent authorities through supervisory review to apply a down-factor, where the firm’s investment in its control environment is such that it poses less risk than a comparable firm which has not invested as heavily in controls.
LGIM agrees with the initial K-factors identified, so long as they produce reasonable risk proxies for calculating capital requirements. There are a number of ways these K-factors could be defined that may provide significant variations in capital requirements. There may also be difficulties in establishing consistent definitions that are capable of being extracted from firms’ administration and financial systems without significant costs. We agree that the K-factors should be based on information readily available to firms rather than being data captured solely for regulatory purposes.
We consider that assets under management (AUM) should be merely a starting point for assessing the level of risk inherent in a firm’s business model, and not a risk in and of itself. It is important that this K-factor is developed in a way which takes account of the different types of products which form part of a firm’s business offering, given that this has a significant impact upon risk profile, rather than just the size and volume of portfolios under management. For example, a firm which has a high AUM but focuses largely on index tracker funds may have a lower risk profile than, say, a firm with a small AUM which offers mostly structured products that are highly leveraged. The capital assessment should therefore drill down deeper into the firm’s business model to identify the composition of its AUM, which may then be weighted to reflect the risks. Without further consideration of such factors, AUM cannot be a suitable and comparable metric for assessing a firm’s risk profile. Furthermore, if AUM is to be introduced as a K-factor, it is essential that a harmonised definition of AUM is introduced across the EU, as well as common industry standards for the underlying calculations.
We believe that the metric of number of orders handled is not sufficient to capture all the relevant considerations in determining the customer orders handled (COH) K-factor. The size and complexity of the orders a firm handles are clearly very important in calculating the level of risk to posed customers, as well as the time-frequency at which they take place. A more holistic approach to COH must be considered if this is to be an effective risk proxy that in turn contributes to an effective capital assessment. There must also be clarity around what constitutes an order; for example, where several consecutive orders are executed with the same counterparty, this may be viewed as a single order if they are interlinked. There are also likely to be differences between firms based on their product mix and the size and nature of the assets traded. In many markets a high volume of trades does not necessarily result in a greater risk of error.
LGIM does not agree that the ‘up lift’ measure on its own is an appropriate way to address the indirect impact of exposure risk which a firm presents. The ‘up lift’ measure pre-supposes that the firm will pose a higher exposure risk to its customers or to markets, and therefore that a higher capital requirement should be applied. In practice, it may be appropriate to apply a ‘down lift’ in the case that a firm poses a lower exposure risk, where its subsequent impact is in fact lower than originally anticipated.
Nevertheless, any ‘up lift’ (or indeed ‘down lift’) factor is unnecessary for firms which do not trade as principal and therefore do not have significant balance sheet exposures. Firms which are not highly leveraged in this way should therefore not be subject to such a requirement. Nevertheless, we accept that an adjustment should be made to the capital numbers calculated via the K-factors, based upon an assessment of the firm’s control environment and its capability to manage within Risk Appetite. Whilst there is likely to be an element of subjectivity in this assessment, we consider this to be an important factor in ensuring that firms are motivated to establish and embed effective risk management practices. To achieve consistency, we propose that such subjectivity could be monitored or indeed set by competent authorities through supervisory review.
LGIM believes a stand-alone approach based solely on the fixed overhead requirement (FOR) would provide a more proportionate methodology to set capital requirements.
In principle LGIM believes that the FOR should remain part of the capital regime; however, we would like further clarity over how the EBA intends to apply the fixed overhead requirement (FOR) to the prudential regulation of investment firms. We agree that the definition of expenditures which are classed as “fixed” should be reviewed. In our view, the current definition of the FOR under CRD IV does not allow sufficient flexibility to deduct avoidable expenditure (such as on project initiatives), or expenditure which is an inherent cost of sale and is linked to the same AUM from which revenue is derived (such as market data costs). For the latter, we believe that excluding such expenditure which is inherently linked to income would be appropriate in ensuring consistency of treatment across Member States. In particular, it would cater for Member States where the industry is moving towards greater transparency in pricing, by requiring asset managers to charge an “all-in” fund management fee and therefore directly paying for fund costs that, in other Member States, may be levied directly on investment funds. This could put investment firms operating under such regimes at a competitive disadvantage, as the impact of the differing practices would be to gross up investment firms’ revenue and expenditures. If such costs are not excluded from the FOR calculation, these investment firms would suffer a comparatively higher requirement.
LGIM does not believe that large firms which trade financial instruments such as derivatives should be considered bank-like. The key considerations should be the extent of the firm’s balance sheet exposure and whether it trades on a principal basis, given that this generates ‘bank-like’ risks to firms, and not merely the fact that it trades financial instruments. However, we agree that additional complexity in a firm’s business operations should be captured by the set of K-factors, so long as this is approached in a proportionate way.
LGIM agrees that the K-factor approach is appropriate for producing a new prudential regime which is based upon the specific risks that investment firms present. In applying the rules, we believe that the individual firm’s risk profile should be properly considered, with onerous requirements dis-applied where the firm is not exposed to a particular risk. For example, the ‘client money held’ K-factor clearly should not be applied universally, given that only a limited number of asset managers hold client assets and therefore pose this risk to customers. The new regime should therefore be adaptable enough to reflect the business models of the diverse range of investment firms operating within the industry.
LGIM views the current CRR approach of using IFRS balance sheet values as the simplest option. However, prudential deductions are currently made which reflect the balance sheet value of deductible items at the reference date, rather than the value at which they were included in “audited reserves”. We have outlined our preference to amend the definition of Own Funds to allow unaudited profits, which would mean that deductions and the value of Common Equity Tier 1 items pre-deductions are aligned.
LGIM’s recommends that the starting point for the definition of “capital” and “Own Funds” should be the current definition in the CRR but with one amendment: the inclusion of unaudited profits within Own Funds. This would ensure that the calculation of capital and capital requirements under the K-factor approach uses the most up-to-date assessment of the firm’s loss absorbency at any point in time. The quality of capital could be maintained by imposing restrictions on setting allowable dividends, with reference to regulatory capital requirements or treating unaudited profits as a separate Tier.
Given the diverse nature of the activities performed by investment firms, there are few (if any) market-related stresses that would be common across all firms. A stress test set with reference to the FOR could be used to replicate a scenario of investment firms operating in wind down scenario without income being generated. In LGIM’s view, this approach to liquidity would be the simplest and most proportionate approach.
An alternative approach could be to apply a common set of haircut requirements to a firm’s liquid asset base and model of the liquidity coverage ratio changes under these stresses. For instance, percentage haircuts could be set for classes of liquid assets, reflecting decreasing levels of liquidity.
LGIM agrees that such an approach would provide a proportionate method of ensuring investment firms retain sufficient access to liquidity. As noted in the discussion paper, the main source of liquidity risk facing investment firms which do not undertake propriety trading is matching operational cash inflows from revenue generating activities with the outflows from the expense base. This would have the benefit of harmonising the approach to liquidity management between investment firms and firms regulated under the AIFMD.
In LGIM’s view liquid assets should include all those assets that can be readily converted to cash in a short (<30 days) timeframe. These would include:
• cash and deposit accounts held with credit institutions;
• investment in government securities;
• investments in securities for which liquid markets exist;
• receivables (provided not overdue and within 30 days of due date) – this would include intragroup receivables where, for instance, the group has a central Treasury function to pool liquid resources; and
• other realisable assets.
If the definition of Own Funds is updated to include current unaudited profits (i.e. the net asset value of the firm less prudential deductions), an approach to defining liquid assets could be to deduct any illiquid assets from Own Funds (similar to the AIFMD approach).
LGIM agrees with the need for a minimum liquidity standard for investment firms, but this must be appropriately linked to wind down requirements which must themselves be properly defined for regulatory purposes. If this is not the case, firms may experience cash shortfalls, affecting their ability to meet liabilities as they fall due and making an orderly wind down impossible.
We note the work done by the Financial Stability Board (FSB) in developing a set of recommendations to address structural vulnerabilities from asset management activities. The proposed recommendations will now form the basis of global industry standards to be produced by the International Organisation of Securities Commissions (IOSCO) on managing liquidity and other systemic risks. As the development of the new prudential regime for investment firms moves forward, it is important that the EBA works in tandem with IOSCO to ensure that, where there is any crossover, a consistent and universal set of standards are applied regarding liquidity management.
LGIM agrees that investment management firms should have a documented policy outlining their exposure and approach to managing liquidity risk.
LGIM considers a reporting scheme for identifying concentration risk to be disproportionate to the level of risk faced by investment firms. For firms without a trading book, the source of large exposures is likely to be cash deposits with banks and credit institutions already subject to robust prudential regulation. As such the level of risk posed by such firms is low. To avoid large exposures, firms could seek to diversify where cash is held by operating separate bank accounts with multiple credit institutions. In practice this is likely to be operationally cumbersome, resulting in increased costs, and could be counterproductive to liquidity management by creating additional complexity to payment and cash management processes.
In addition, such a regime could be detrimental to Class 3 investment firms as they may have a concentrated client base, putting them at a disadvantage as they attempt to gain scale.
Finally, LGIM considers that any such reporting should exclude intra-group exposures, where cash can be readily transferred from one member of the group to another without impediment.
LGIM agrees with the outlined approach in the discussion paper and considers that the EBA’s suggestion of codifying a common set of derogation criteria into the regime, rather than relying on the (potentially divergent) judgement of competent authorities, to be the best option.
We note that the discussion paper provides for the expected derogation of consolidated supervision of an investment sub-group of firms already included in a banking consolidation. We do, however, consider that derogation of consolidated supervision should apply equally to investment firms included in the consolidated supervision of insurance groups under Solvency II.
LGIM recognises that, if the implementation process is not applied proportionately, investment firms which form part of a group may become subject to two sets of prudential requirements, leading to an unduly complex regime. Similar to banking groups subject to the CRD regime, we have some concerns around how investment firms which form part of insurance groups subject to Solvency II will be treated, or as with LGIM, where the Group offers investment services via an insurance contract. In these instances the legal entity will be subject to two very different capital regimes for similar business models, and so there is a risk that this would make day-to-day business management more challenging.
In practice, group structures can be very complex, with a vast range of legal entities, and this must be taken into account when applying prudential standards. We believe that the new regime should be applied at the individual firm level, with the removal of any alternative capital requirements which emanate from group-wide prudential regulatory frameworks, where the overall Group is subject to comparable consolidated supervision (such as under Solvency II or CRD IV).
LGIM believes that any obligations arising from the management of pension client assets should be considered when assessing the prudential measures to be applied to individual firms. As noted above, competent authorities should consider the governance arrangements of investment firms in the context of their group structures. Where similar governance arrangements exist under alternative prudential regulation at the group level, we see little benefit in duplicating such requirements at the solo entity level.
LGIM has found the regulatory reporting requirements under the Pillar 3 disclosure particularly burdensome. In particular, the need to identify, classify and risk weight all balance sheet counterparties when assessing credit risk under Pillar 1, even though credit risk forms a minimal part of the firm’s Pillar 1 capital requirement, has been very challenging. In our view, the mandatory templates prescribed by the EBA for Pillar 3 reporting add little value and are potentially confusing to users without a regulatory background. The objective of transparency of reporting could be better achieved by setting minimum reporting standards (for instance, the value of Own Funds, capital requirements and the capital surplus), which would simplify the reporting documents and allow their publication within financial statements more readily.
LGIM believes that the level of planning set out in the BRRD would be disproportionate for most investment firms. An alternative approach, given the risk profile and prudential objective of ensuring an orderly wind down, could be to require investment firms to have a documented wind down plan only. It is essential that any additional prudential tools are proportionate.
LGIM supports the maximum harmonisation possible in relation to the governance requirements across the various regulatory initiatives. Regarding remuneration, very similar requirements are contained in the CRD, AIFMD, UCITS V and Solvency II, leading to the application of several different remuneration codes to achieve essentially the same objectives. We therefore recommend that any new remuneration code is aligned with the requirements under existing regulatory initiatives.
LGIM’s agrees that, for agency-based asset managers, this is a more sensible approach to prudential management. The primary risks for most asset managers remain operational; the main consequence that firms need to be capitalised for is to ensure that customers are protected from the impact of operational risk events.
LGIM believes that remuneration requirements should reflect the risk profiles, business models and pay structures of investment firms, and seek to reward adherence to a robust risk control framework as well as strong investment performance. We support the EBA’s view that a separate regime for investment firms other than ‘systemic and bank-like’ is appropriate, provided that such a regime is proportionate in terms of governance and disclosure obligations.
LGIM’s preference is to have a separate regime for investment firms; we do not support the approach of amending the current CRR regime to take account of a higher degree of proportionality. The business models of banks and asset managers are sufficiently dissimilar to warrant distinct prudential regimes for each sector. Any attempt to adapt the CRR regime is unlikely to resolve the issues in current prudential regulation of investment firms – i.e. that the risk proxies which form the basis of the capital requirements reflect the risks faced by banks, and not investment firms. We therefore believe that the best solution is to create a tailored regime which includes the individual risks faced by investment firms, based upon both market analysis and industry feedback from such firms.
LGIM concurs with the problems highlighted for investment firms in the discussion paper. We have found the operational burden of calculating and disclosing credit exposures to be particularly problematic aspects of the current regime.
In addition, the definition of “Own Funds” under CRD IV is such that it provides a retrospective view of the investment firms’ capital positions at the point their audit opinions were signed. This means that the balance sheets of firms subject to audit once a year can bear little resemblance to Own Funds reported to national competent authorities.
There are also issues around the model-based approach to prudential regulation under the CRD regime. The application of such tests within firms is very complex, and so senior management may be unable to fully appreciate the impact and understand the way in which capital requirements are calculated. This confusion may create insufficient oversight and monitoring of the level of capital held by firms.