ICI Global Comment Letter on ESMA Consultation on Guidelines for Regulating Alternative Investment Fund Leverage (pdf)

ICI Global Response to ESMA Consultation on Guidelines on Article 25 of Directive 2011/61/EU ICI Global1 appreciates the opportunity to comment on ESMA’s proposed guidelines to the National Competent Authorities (“NCAs”) on the assessment of alternative investment fund (“AIF”) “leverage-related systemic risk” and on the design, calibration and implementation of AIF leverage limits.2 e proposed guidelines represent ESMA’s response to recommendations from the European Systemic Risk Board to operationalise macroprudential leverage limits for AIFs.3 Two-step assessment process (questions 1-6) ESMA’s proposed guidelines on AIF leverage-related risk assessments pertain to NCA assessments in accordance with Article 25 of the Alternative Investment Fund Managers Directive 2011/61/EU (“AIFMD”), meaning that the goal of the assessments is to identify risks from AIF leverage of a magnitude that would negatively impact the financial system or result in disorderly markets.4 It is therefore critical that NCA assessments focus squarely on funds that are more likely to pose potential risk, while appropriately excluding from further evaluation most funds that do not. In its guidance to NCAs, ESMA should closely follow the approach set forth in the International Organization of Securities Commissions’ (“IOSCO”) Recommendations for a Framework Assessing Leverage in Investments Funds (“IOSCO Framework”).5 Following a request from the 1 ICI Global carries out the international work of the Investment Company Institute, the leading association representing regulated funds globally. ICI’s membership includes regulated funds publicly offered to investors in jurisdictions worldwide, with total assets of US$31.3 trillion. ICI seeks to encourage adherence to high ethical standards, promote public understanding, and otherwise advance the interests of regulated investment funds, their managers, and investors. ICI Global has offices in London, Hong Kong, and Washington, DC. 2 See ESMA, Consultation Paper: Guidelines on Article 25 of Directive 2011/61/EU (27 March 2020) (“Consultation”), available at https://www.esma.europa.eu/sites/default/files/library/esma34-39- 967_consultation_paper_on_guidelines_on_art_25_aifmd.pdf. 3 See ESRB, Official Journal of the European Union, Recommendation of the European Systemic Risk Board of 7 December 2017 on liquidity and leverage risks in investment funds (ESRB/2017/6) (4 April 2018), available at https://www.esrb.europa.eu/pub/pdf/recommendations/esrb.recommendation180214_ESRB_2017_6.en.pdf. 4 See EU, Official Journal of the European Union, Directive 2011/61/EU of the European Parliament and of the Council (8 June 2011) at Article 25(1) (1 July 2011), available at https://eur-lex.europa.eu/legal- content/EN/TXT/PDF/?uri=CELEX:32011L0061&from=EN. 5 See IOSCO, Recommendations for a Framework Assessing Leverage in Investments Funds (Dec. 2019), available at https://www.iosco.org/library/pubdocs/pdf/IOSCOPD645.pdf. ICI Global | Page 2 of 7 Financial Stability Board,6 the IOSCO Framework was designed to achieve a substantially similar goal – to facilitate monitoring of leverage in investment funds that could potentially pose risks to financial stability. In deriving its framework, IOSCO spent considerable effort on determining how to measure investment fund leverage for financial stability purposes. It is therefore not necessary for ESMA to take a more expansive approach. In our responses below, we recommend targeted changes to the assessment process to help NCAs monitor more meaningfully leverage for financial stability purposes. Leverage limits (questions 7-9) ESMA’s proposed guidelines appear to contemplate that NCAs should make use of the broad authority provided in Article 25 of the AIFMD to impose leverage and “other” limits on funds that they identify as posing systemic risk. In particular, the guidelines suggest that NCAs may choose to impose limits on the proportion of certain assets in which AIFs may invest or implement redemption policies and restrictions on AIFs. We believe that NCAs should prudently exercise this broad authority to avoid market disruption and harm to investors. Fund managers have used investments that create structural (e.g., debt) and synthetic (e.g., derivatives) leverage successfully for decades to, among other things, hedge risk, manage duration, enhance liquidity, and gain or reduce exposures more efficiently. Imposing broad limits, such as these, has every potential to eliminate or substantially restrict managers from using these important tools for the benefit of a fund and its investors. To help avoid unintended consequences, we recommend that NCAs consider consulting on possible limitations that they may impose to receive appropriate feedback (and to help funds understand the tools that could potentially be used) and to consult confidentially with any funds before they become subject to such limitations. Q1. What are your views on the equency at which the risk assessments should be performed by NCAs? We do not object to ESMA’s proposed guidelines recommending that NCAs perform leverage- related risk assessments on AIFs on a quarterly basis. AIFs generally must report relevant leverage- related data quarterly, and NCAs could review that data consistent with those timeframes. ESMA should confirm that the guidelines would not recommend that AIFs that report on a less than quarterly basis to increase the frequency of their reporting. Q2. What are your views on the sample of funds to be included under Step 1? Do you agree in including in the risk assessment not only substantially leveraged funds but also funds not employing leverage on a substantial basis which may pose financial stability risks? 6 See FSB, Policy Recommendations to Address Structural Vulnerabilities from Asset Management Activities (12 January 2017) (“FSB Policy Recommendation”), available at http://www.fsb.org/wp-content/uploads/FSB-Policy- Recommendationson-Asset-Management-Structural-Vulnerabilities.pdf. Recommendation 10 of the FSB Policy Recommendation asks IOSCO to “identify and or develop consistent measures of leverage in funds to facilitate more meaningful monitoring of leverage for financial stability purposes, and help enable direct comparisons across funds at a global level.” ICI Global | Page 3 of 7 We strongly believe that the sample of funds to be included in Step 1 is overinclusive. We also disagree with broadly including funds that do not employ substantial leverage in this leverage-related risk assessment. ESMA’s proposed guidelines recommend that, under Step 1, NCAs identify three categories of AIFs to be assessed under Step 2: a) AIFs employing leverage on a substantial basis compared to their net assets; b) AIFs not employing leverage on a substantial basis but whose regulatory assets under management exceed EUR 500 million at the reporting date (which we refer to in this response as the “large fund/little leverage” standard); and c) other AIFs whose “unusually high use of leverage”—as compared to other funds—may pose risks to financial stability.7 e proposed Step 1 guidelines would capture many funds that do not pose financial stability risk and differ from the IOSCO Framework. NCAs’ Step 1 analysis, instead, should serve as an effective screening tool to exclude from any further consideration funds that are unlikely to pose risks to the financial system. To be effective, the step should be consistent with IOSCO’s approach and meaningfully narrow the group of funds that will be subject to further regulatory analysis. In this regard, ESMA should modify the Step 1 guidelines to: i) exclude the “large fund/little leverage” standard to efficiently remove from further consideration funds whose leverage use does not raise systemic risk concerns; ii) exclude the “unusually high use of leverage” standard to eliminate comparisons that could misidentify for further analysis many funds that do not raise systemic risk concerns; and iii) recommend NCAs analyze fund leverage by broad asset class and long and short positions to better reflect the fund’s risk to the financial markets. Exclude the “large fund/little leverage” standard. ESMA’s proposed Step 1 approach would single out for further analysis funds that are large (funds whose regulatory assets under management are greater than EUR 500 million) but that do not employ leverage on a substantial basis. Fund size without more, however, provides little indication of leverage-related systemic risk. Exclude the “unusually high use of leverage” standard. ESMA’s proposed Step 1 guidelines would cause NCAs to deem a fund to engage in an “unusually high use of leverage” when it uses more leverage than other similar AIFs or its own historical median or average leverage value. Employing this standard could create a false impression of leverage use. For example, a fund that historically has not used leverage but that determines to use leverage during a period may be flagged for further review under such a standard, even if the fund’s overall leverage use is minimal. On the other hand, a fund that does not use as much leverage overall, but applies all of its leverage to riskier underlying assets, might not be flagged for further review. To avoid these types of “false positives” and “false negatives” as part of the Step 1 analysis, ESMA should eliminate the standard comparing a fund to other funds in its category or to its prior leverage usage without consideration of the magnitude of such leverage. 7 ESMA clarifies that “unusually high use of leverage” is use of leverage that differs significantly (e.g., a high percentile in the distribution) from that of other funds by comparing the AIF’s leverage value with: 1) the median or average value of leverage of AIFs of the same type (e.g., hedge fund, private equity, real estate, fund of funds, other AIFs); and 2) the AIF’s historical median or average leverage value. ICI Global | Page 4 of 7 Recommend NCAs evaluate a fund by asset class and long and short position. Under ESMA’s Step 1 analysis, NCAs could consider offsetting and hedging transactions through its use of the commitment approach as an indicator.8 We believe that NCAs must consider offsetting and hedging transactions when evaluating systemic risk, otherwise many funds that engage in leverage- related investments to hedge their portfolios may be unnecessarily identified for further evaluation. e proposed Step 1 analysis would fail to provide sufficient details about the fund’s impact on underlying asset classes and markets to which a fund might gain net leverage exposure. Without this key information, NCAs will have difficulty conducting a robust analysis on a fund. Similar to the IOSCO Framework, NCAs should evaluate each fund considering the fund’s exposure to various broad asset categories (e.g., equity derivatives, equity securities, fixed income securities, credit derivatives), broken out by long and short positions. In a recent letter, ESMA confirmed the appropriateness of this approach as “a means of efficiently identifying those funds that are more likely to pose risks to the financial system.”9 Evaluating this information separated by broad asset categories provides more meaningful insight than simply evaluating one combined figure, such as under the commitment approach. It would allow NCAs to better identify funds of interest and their relevant exposures, enabling them to see a fund’s basic asset allocations and exposures to higher risk assets along with the directionality of their positions. is knowledge is crucial because different asset classes have differing levels of risk. Q3. Do you agree with the proposed leverage and size threshold identified under Step 1? Would you set the same threshold for all AIFs, or would you be in favour of setting different thresholds based for different types of AIFs (e.g., real estate, hedge funds, private equity, etc.) or sub-types of AIFs (please specify) based on a statistical analysis (e.g., percentile)? Should you prefer the latter option, please provide proposals and detailed arguments and justification supporting them. We cannot evaluate the proposed leverage ratio and size thresholds identified under Step 1 because it is unclear how those thresholds relate to systemic risk. e Consultation does not explain the rationale or justification for those thresholds or at what leverage thresholds funds could pose risk to the financial system. NCAs should establish jurisdictional thresholds based on the impact that a fund’s leverage might have on the NCA’s local markets. ESMA, likewise, should demonstrate how the proposed thresholds might impact the global markets. We are not in favor of setting different thresholds based on AIF type. Regardless of their type (e.g., hedge fund, fund-of-funds), many AIFs can invest in different asset categories and markets.10 8 Under the commitment approach, funds may net derivatives and other investments when they are concluded with the sole aim of eliminating the risks linked to positions taken through the other investments and only when the trades offset the risks linked to the investments, leaving no material residual risk. Funds may exclude hedging transactions when they are concluded with the sole aim of offsetting risks linked to positions taken through the other investments and only when a number of conditions are satisfied (e.g., there must be a verifiable reduction of risk and the transactions must be in the same asset class). 9 See, e.g., Letter from Steven Maijoor, Chairman of the Management Board of the European Securities and Markets Authority, to Valdis Dombrovskis, Executive Vice President for an Economy that Works for People, European Commission, dated 18 August 2020, at 9, available at https://www.esma.europa.eu/press-news/esma-news/esma- recommends-priority-topics-in-aifmd-review. 10 Of course, certain AIF types may limit the AIF’s investments to certain asset categories or markets (e.g., real estate or private equity funds). ICI Global | Page 5 of 7 erefore, setting different thresholds based on AIF type would not identify funds that could pose risk to any particular asset class or financial market and, accordingly, would yield little additional benefit. Further, we understand that classifying AIFs by type for reporting purposes has proven difficult, and many AIFs may fall into the “other” category. We also object to NCAs establishing Step 1 thresholds based on statistical analysis. As we highlighted in response to Question 2, statistical analysis broadly comparing AIFs to other AIFs generally or in particular categories would create a number of misleading “false positives” and “false negatives.” Q4. Would you identify other relevant transmission channels? No. Q5. What are your views on using not only leverage indicators but also other types of indicators, such as those indicated under Table 2 of the dra Guidelines? Do you agree with the list of indicators provided? We do not object to NCAs considering other indicators (e.g., those listed in Table 2 of the proposed guidelines) in addition to leverage indicators to assess whether a fund could pose any of the identified systemic risks. Additional indicators can provide more information and/or context to consider, and NCAs should factor in as many pieces of relevant available information as possible to exclude funds from further analysis. In assessing indicators, the NCA should explain how the particular indicator ties to financial stability. For example, when evaluating market impact, NCAs should show not only how a fund’s use of leverage could materially affect the operations or performance of a given market but at what levels, given the size of the market. ESMA also should reiterate that the goal is to assess whether a fund’s leverage contributes to systemic risk. erefore, any indicator must directly tie to how the fund’s use of leverage could create financial stability risk. Certain Table 2 indicators apply to groups of AIFs. As further described in response to Question 8, the extent of the term “groups” in these contexts is open-ended and ambiguous. All funds are part of some group. We therefore do not agree with the recommendation that NCAs assess and impose limits on groups of funds. If, however, an NCA intends to assess a group of funds, it must clarify the extent of the group and the rationale for singling the group out for this leverage-related assessment. Further, NCAs should ensure that their assessments of whether a fund is more likely to create financial stability risk based on these and other indicators are objective and consistent. ey should be grounded on a reasonable system of evaluation that is reproduceable. In this regard, NCAs should establish clear standards for how they will make assessments, and these standards should be publicly communicated. Q6. What are your views on using not only AIFMD data but also other external data sources to perform the assessment? Which types of external data sources would you consider more useful for the purpose of performing the assessment under Step 2, other than those already identified in the Annex to the dra Guidelines? ICI Global | Page 6 of 7 We believe that the AIFMD data should give NCAs much information to evaluate the potential systemic risk an AIF may provide. We do not suggest any further types of external data sources for NCAs to consider for purposes of performing the Step 2 assessment, other than those ESMA already identifies. In evaluating external data sources, we encourage NCAs to eschew the temptation of placing too much emphasis on any one data point. Q7. Which other restrictions would you consider as appropriate? None. Before NCAs implement any restrictions, we strongly encourage them to engage actively with the identified funds and their managers to address questions and issues. Q8. What are your views on the application of the leverage limits? Should those be applied only on the single fund or, where appropriate, limits should also be applied on group of funds? In this case, how would you identify the group of funds? We agree with providing NCAs, which are in the best position to evaluate fund leverage in their jurisdiction, the discretion to derive and impose fund leverage limits. ESMA, however, should recommend that NCAs narrowly tailor any leverage limits and ensure that the limits are implemented transparently through a consultation process. In addition, NCAs should apply the leverage limits only to single funds. e Consultation broadly states that NCAs should impose leverage limits on AIFs posing risks to financial stability (i.e., those identified under the assessment phase). While the proposed guidelines provide direction on the length and the level of the leverage limits, ESMA also should encourage NCAs to apply any leverage limits narrowly to address the risks identified by the NCA in the assessment phase (e.g., risk of market impact, risk of direct spill-overs to financial institutions, and risk of interruption in direct credit intermediation). Leverage limits have every possibility of creating a procyclical effect. For example, imposing limits when risks are increasing may cause funds to deleverage at a time when a market segment is declining. Leverage limits also could restrict the use of efficient portfolio tools that ultimately benefit fund shareholders. erefore, limits should not be used as some supposedly prophylactic measure without clear, detailed evidence that they are necessary and appropriate. Before implementing the restrictions, therefore, ESMA should recommend that NCAs consider consulting on the possible limitations that they might impose (including both leverage limits and “other” restrictions, their length and amount, and when those limits might be imposed), provide an analysis demonstrating the cost/benefit impact the limits could have on the affected funds, and seek feedback. In addition, as we noted in response to Question 7, ESMA should recommend that NCAs consider consulting confidentially funds subject to such limitations before they are imposed. Consistent with this comment, ESMA should recommend that NCAs clearly justify the length and the amount of the limits at the time of implementation. In this regard, the Consultation states that fund leverage limits should be imposed temporarily to limit the buildup of risk, such as when they contribute to excessive credit growth or the formation of excessive asset prices. It is unclear how an NCA will make either of those determinations and when each returns to “normal” conditions. NCAs should explain how these risks will be evaluated and how leverage limits of a set duration and amount are intended to ameliorate the risks. ICI Global | Page 7 of 7 Leverage limits should apply only to a single fund. As discussed in response to Question 5, all funds are part of some group, and the term “group” is unclear and undefined. erefore, its meaning is subject to different interpretations by different NCAs, which could lead to inconsistencies across jurisdictions. Accordingly, we strongly recommend eliminating classifications and leverage limits by group, which also appear to be inconsistent with the IOSCO Framework. If, however, leverage limits are imposed on a group of funds, NCAs must clarify what that group is and why it is being singled out. ere needs to be a rational, publicly demonstrated reason for implementing the limit on a group, justifying the length and amount of the limit. In addition, NCAs must communicate regarding their interpretations of its meaning to ensure consistency across jurisdictions. Q9. How would you assess the efficiency of leverage limits in mitigating excessive leverage? We generally agree with ESMA’s proposed standards for evaluating the efficiency of leverage limits. Importantly, NCAs must consider whether the leverage limits are proportional to the systemic risks an AIF may pose “in order to ensure that the sector remains able to provide valuable services to the economy.” In conducting this evaluation, NCAs should clearly identify the risks that it intended to mitigate, as we recommended in response to Question 8, and rigorously assess if the imposed leverage limits are working as intended to mitigate those risks.