EFAMA REPLY TO THE EBA / ESMA CONSULTATION PAPER FOR BENCHMARKS SETTING PROCESSES IN THE EU EFAMA 1 welcomes the opportunity to provide comments on the EBA / ESMA joint consultation paper on benchmarks setting processes in the EU. EFAMA is the representative association for the European investment management industry. EFAMA represents through its 27 member associations and 60 corporate members about EUR 14 trillion in assets under management of which EUR 8.7 trillion managed by 54,000 investment funds at end September 2012. Just under 36,000 of these funds were UCITS (Undertakings for Collective Investments in Transferable Securities) funds. EFAMA members are in principle not producing or contributing to data on which the calculation of benchmarks is based. Asset managers are usually users of benchmarks and financial and other types of indices when managing portfolios on behalf of their clients, either by tracking benchmarks and market indices directly or as a reference to compare across different investment opportunities. Following the recent Libor scandal, restoring market credibility and re establishing confidence in benchmarks should be the first and foremost priority. We therefore fully support the inclusion of cases of benchmarks manipulation into the market abuse regime. Further on EFAMA shares the goal of this joint consultation paper to further assess the type of principles and if considered appropriate of regulatory framework to improve the governance of benchmarks and guarantee high transparency, responsibility and accountability standards. We would first like to provide you with a few general comments before answering the specific questions raised in the Consultation Paper. A. General Remarks 1. The aim of the recent consultation issued in November 2012 by the EU Commission as well as of the present joint EBA/ESMA consultation paper is to assess the need for further principles and regulation regarding the use of benchmarks in financial transactions. Therefore it seems clear that the main target is the submission and administration of benchmarks and not directly the use of those benchmarks by market participants. The later will become relevant only as a 1 For more information about EFAMA, please visit www.efama.org. rue Montoyer 47, B 1000 Bruxelles +32 2 513 39 69 Fax +32 2 513 26 43 e mail :info@efama.org www.efama.org VAT Nr BE 0446.651.445
2 complementary condition where necessary to strengthen the transparency, responsibility and accountability standards. Market participants are therefore free to select benchmarks or indices according to the needs of their clients as long as indexes meet regulatory requirements. 2. ESMA has recently published on the 18th of December 2012 its Guidelines on ETFs and other UICTS issues, where a set of conditions for the use of financial indices as benchmarks by UCITS is included. Given however the lack of similar guidelines for other benchmark users this may cause UCITS managers significant disadvantages compared to other market participants. This stresses even more that in order to ensure a level playing field for all market participants and benchmark users, the starting point of any set of principles and/or regulatory framework should be the submission, calculation and administration of benchmarks rather than their use by a particular group of market participants. Any new set of principles for benchmark setting processes should also be consistent with the already applied principles on the use of indexes in the case of UCITS. 3. Given the extensive list of variable benchmarks and market indices, regulatory priority should be given to benchmarks such as LIBOR and EURIBOR. This was indeed the case in dealing with the LIBOR scandal so far where malpractices and market manipulations called for an urgent action. Restoring market credibility should therefore start from those types of benchmarks which are based on subjective estimates and are the key elements of the interest rate swaps market and the floating rate loans. 4. Apart from the abovementioned types of benchmarks, any further regulatory initiative in the field of market indices which are based on objective data and transaction volumes should be fully and carefully calibrated taking into consideration the additional costs and risks that may arise. In the case of commercial based indices, asset managers as users and their clients as end investors are already assuming important compliance costs based on the existing regulatory requirements. Therefore any additional regulatory framework that will result to even higher compliance costs should be put in place only when considered necessary to address market disruptions. In that regard a cost benefit analysis should take place prior to the introduction of any new principles or regulatory requirements, assessing the risks and benefits of a new regulatory regime vs. an enhanced framework of transparency and disclosures. 5. Moreover a one size fits all regulatory approach is not appropriate for the wide scope of benchmarks considered in this consultation paper. The benchmarks covered present significant differentiations regarding the underlying markets upon which they are based, their use, economic impact, susceptibility to manipulation and systemic relevance. In this regard we consider that the criteria for regulatory distinctions foreseen in the current IOSCO Consultation report on Financial Benchmarks to be a good starting point for a more targeted regulatory approach. This could consist on stricter regulatory regime for some types of benchmarks and less restrictive for others where timely transparency of index prices, constituents and weightings will be considered sufficient.
3 6. Changes proposed to a benchmark s framework should take fully into account the potential risks of market disruption and dislocation. Minimizing such risks as well as ensuring the continuity of financial instruments and the stability of the funds that reference such benchmarks should be key priorities. 7. As indices and benchmarks operate globally and are located in a number of different regions, discussions and work at international level mainly under IOSCO s initiative to define global principles for the setting of benchmarks should form the basis for any decision at the EU level. This will ensure legal clarity on the application of the EU rules and prevent disadvantages for the EU benchmarks providers and users. EBA and ESMA should also clarify the use of third country benchmarking. B. Answers to questions raised in the Consultation Paper Question 1: Definition of the activities of benchmark setting Do you agree with the definitions provided in this sector? Is this list of activities complete and accurate? As a general remark we would like to stress that the term benchmark as used in this consultation paper corresponds mostly to the notion of index. Indices are used for the calculation of the amount payable under a financial instrument, for the pricing of an instrument or as a reference to measure market performance. The third case is the case where an index is used as a benchmark. The definition on benchmarks under (i) is covering the two first of the abovementioned cases of index and correctly does not refer to the case when an index/benchmark is used as a reference to measure market performance. However, due to the use of term benchmark instead of index it should be made explicit that this term does not cover the case of an index that is used as a reference standard to measure market performance. Moreover, a clear distinction should be added to differentiate between benchmarks used by more than one or not a particular group of investors and the so called customized/ bespoke benchmarks. Additional definition could be: The term benchmark under this consultation paper does not refer to customized/bespoke benchmarks which are used as a combination and/or modification of existing public indices on the request of one or a very limited number of market participants and individually agreed according to particular investment objectives and strategies. Customized/bespoken benchmarks are used on individual purposes based on existing ones and should in no case be considered as new type of benchmarks. Asset managers using these indices are neither producing nor administrating new benchmarks. They apply existing ones and/or combination of them according to the particular needs and requests of their clients.
4 Given that such individually agreed benchmarks pose significantly lower systemic risks and implement specific only risk/return profile, inserting this additional criteria will enable differentiated regulatory approach or if appropriate their exclusion from the scope of any future legislative framework. Regarding the definition under (x) on benchmark users, a further clarification should be added to make it explicit that market participants that use benchmarks as a reference to compare across different investment opportunities are not to be considered as users for the scope of this consultation. Question 2: Principles for benchmarks Would you consider a set of principles a useful framework for guiding benchmark setting activities until a possible formal regulatory and supervisory framework has been established in the EU? EFAMA considers the steps taken so far, i.e. including the market indices in the new market abuse regime and the EU Commission s consultation in order to assess the need for further regulatory measures, as the right ones towards restoring investors confidence in the market and improving financial stability. At the same time benchmarking setting processes is a global operation. Therefore, the on going IOSCO discussions on defining global principles for the setting of benchmarks are the essential points on which EU legislators and policy makers should base any future decisions. On the regulatory approach we consider that due to the wide range of benchmarks covered by this consultation paper a set of priorities should be drawn. We believe regulatory priority should be given to benchmarks which are based on panel decisions and subjective assessments such as EURIBOR and LIBOR. On the other types of indices that are mainly based on transaction volumes and concrete data, any decision for further regulating their setting processes should be assessed against the risks and additional costs it may bring. Commercial indices present significantly fewer opportunities for market manipulation as they are based on objective data and public rules and both providers and users have an increased interest in preserving their accuracy, integrity and quality. The main priority here should be that no regulatory requirement is imposed that without bringing a concrete added value will have unintended consequences in particular on the costs that will eventually be carried from benchmarks providers onto users and end investors. Asset managers are already called to pay for multiple license fees in complying with the existing regulatory requirements. Any additional regulatory oversight will most probably further increase costs for them and ultimately the end investors and at the same time incentivize index providers to maintain the existing oligopoly on fees pricing.
5 Question 3: General principles for benchmarks Do you agree with the principles cited in this section? Would you add or change any of the principles? A.1 Methodology EFAMA supports the need for representative, able to be replicated, measured and where feasible liquid benchmarks. In the case of the liquidity it should be considered on a relative basis as the adequacy of this criteria depends on the type of the underlying assets and the market targeted. For instance liquidity is a very important precondition for benchmarks such as LIBOR and EURIBOR whereas in the case of high yield bonds it would be less adequate. Alternatively to targeting liquidity, another option would be to request transparency as to the methodology. A.2 Governance Structure Transparent governance is a pivotal principle. Measures such as independent audit and sanctions similar to the ones in the market abuse regime as well as consultation with the end user of the relevant benchmark are amongst the main safeguards for achieving transparent governance structures. A.3 Supervision With respect to the supervision, it should be clear that authorities based on government bodies are not the only option. It is important to fully assess all possibilities prior to setting the supervisory principals, amongst which self regulatory bodies run by professionals. A.4 Transparency Transparency should be considered as an integral part of the benchmark setting process. A code of best practices for benchmark providers would allow a more accurate selection of benchmarks and higher consistency regarding investment strategies. Publication on a daily basis of benchmark prices and weightings along with their constituent and structure parts can have a key role for re establishing market confidence on benchmarks. However there are both commercial and systemic reasons that can justify publication of the data with a modest time lag, especially in the cases where delaying publication of data can ensure a limited mitigated impact in periods of market crisis. Regulators and market participants have to jointly decide on the sufficient level of transparency.
6 Particular attention should also be paid to the costs of transparency in particular the costs for benchmark users getting access and license to use benchmarks. Asset managers are called to pay data licenses to access many benchmarks. On top of that they are called to pay for a license in order to be allowed to calculate the performance of the index from the raw data and an extra license to transfer information to their investors as required by the UCITS Key Investor Information Document (KIID) and the recently published ESMA guidelines on ETFs and other UCITS issues. Finally, as the KIID is published regularly on the internet, an additional publication license may apply. It is unacceptable that asset managers are being called to pay multiple times for license fees corresponding to the use of the same index. In that way regulatory requirements are being perceived as costly business opportunities by benchmark providers and incentives for oligopoly pricing and innovative market pricing. This issue should be fully addressed by any future set of principles or regulatory framework by taking a reasonable commercial cost approach as presented in the EU Commission consultation paper (non discriminatory access to and obligation to license use of indices on a reasonable commercial basis). A.5 Transition / Continuity In principle the choice of new or alternative benchmarks should be left to users and their clients based on their analysis on which benchmarks better suit their needs. However, as benchmarks might become less representative due to market evolution, it is important to keep innovating new benchmarks as to make sure they are still representative. Stress resilience and market representation of the index or benchmark are essential characteristics. In the case of changes to existing indexes and benchmarks or transition to new ones particular attention should be put to ensuring minimum disruptions and time flexibility for market participants to reshape contractual arrangements. Replacement should be determined by asset managers or their clients. Any ad hoc change in a very short time which can disturb legal clarity should be avoided. A well defined transition period and regulatory safeguards that will protect legal validity of existing contracts are important. If considered appropriate a clause can be included whereby counterparties agree to either close the contract or change reference if necessary. Question 5: Principles for benchmark administrators Do you agree with the principles cited in this section? Would you add or change any of the principles? We would also see it appropriate to further assess the possibility of an external oversight committee as foreseen in section C.2 of the IOSCO consultation, which will complement individual benchmark administrators in identifying, mitigating and managing potential conflicts of interests.
7 Another overarching principle not mentioned is the full and ultimate responsibility of the administrator, even when it relies on outside parties for different functions. Question 8: Principles for users of benchmarks Do you agree with the principles cited in this section? Would you add or change any of the principles? F1, F2 and F4. The responsibility for suitability, appropriateness and relevance of a benchmark can take place at different levels of setting process, from the submission and calculation of a benchmark until its application by a user. However the main responsibility for the developing, calculating and disseminating of a benchmark should remain on the contributors, administrators and calculators of the data. Asset managers as part of their portfolio management carry validation exercises and monitor the data used to construct benchmarks to the extent they dispose the data available to proceed with the validation. This is for instance the case in the equity and fixed income indices but is not feasible in benchmarks based on panel decisions such as LIBOR and EURIBOR. Moreover even in the case of market indices, validation by users is restrictive as they don t have all data necessary to fully verify the construction or the accuracy of a benchmark. Although they employee some techniques to help identify the appropriateness of a benchmark for an investor, they cannot guarantee it. Therefore the principles in that section imposing requirements on benchmark users not only to regularly assess and notify the administrators on any irregularities they might find, but also to verify the appropriateness of benchmarks and to ensure that the benchmark administrator/calculation agent comply with various principles are not realistic. Not only do they place an inappropriately high burden on users, but they are also required to assume the role of policy makers and supervisors. Moreover if that would be the case, the liability for a false calculation or submission will unevitably cover also users although they have no say on the setting processes of benchmarks. Further on, when the benchmark administrator or benchmark calculation agent is located outside the EU it would be difficult (or impossible) for the user to comply with principle F.2. Moreover asset managers are not always the ones choosing the indices of the portfolios they manage. In several cases it will be the end users who will make this choice according their investment needs and therefore any review decision regarding benchmarks can only be taken either by them or in coordination with them. F3. On the other hand developing robust contingencies in the case a benchmark becomes unavailable as foreseen in principle F.3, is a preventive task that benchmark users are in the position to carry in order to ensure continuity of contracts, but clear guidance is necessary.
8 Question 10: Continuity of benchmarks Which principles/criteria would you consider necessary to be established for the continuity of benchmarks in case of a change to the framework? Please see our response in question 3 transition / continuity issues. Brussels, 15 February 2013 [13 4008]