COMMISSION STAFF WORKING DOCUMENT EXECUTIVE SUMMARY OF THE IMPACT ASSESSMENT Accompanying the document Proposal for a Regulation of the European Parliament and of the Council on indices used as benchmarks in financial instruments and financial contracts
/* SWD/2013/0337 final */
|Bilingual display: BG CS DA DE EL EN ES ET FI FR GA HR HU IT LT LV MT NL PL PT RO SK SL SV|
An index is a statistical measure, typically of a price or quantity, calculated from a representative set of underlying data: when this index is then used as a reference price for a financial instrument or other financial contract it becomes a benchmark. A wide variety of benchmarks are currently produced. These benchmarks differ not only in the underlying data used, but also in the methods used to collect the data, the calculation of the index and their ultimate use.
The benchmark industry generates around EUR 2 billion in revenues for financial and commodity benchmark administrators world-wide. Estimates suggest that the size of the markets impacted could be over EUR 1,000 trillion. The recent alleged manipulation of the interest rate benchmarks LIBOR and EURIBOR has highlighted both the importance of benchmarks and their vulnerabilities. The integrity of benchmarks is critical to the pricing of many financial instruments, such as interest rate swaps and forward rate agreements, and commercial and non-commercial contracts, such as supply agreements, loans and mortgages. They also play an important role in risk management.
The integrity of benchmarks was the focus of the Wheatley Review of LIBOR and the subsequent regulation of LIBOR in the United Kingdom, as well as of the regulation of CIBOR in Denmark. The EBA and ESMA published Principles for Benchmarks-Setting Processes in Europe on 6 June 2013 that provide guidance to participants in the gap before any Commission initiative comes into force. The EBA also carried out a review of Euribor and published recommendations to the EBF-Euribor in January 2013. In addition, the European Commission is undertaking investigations into a possible cartel in relation to EURIBOR and into a potential case of collusion by contributors to price assessments for oil and biofuels by some commodity price reporting agencies.
The FSB is coordinating the international approach on benchmark reform. At its request, IOSCO published Principles for Financial Benchmarks in July 2013. Previously, IOSCO had published Principles on the oversight of oil price reporting agencies in October 2012. Further work is being conducted at FSB, G20 and BIS level.
2. problem definition 2.1 Risk of benchmark manipulation
Benchmarks directly affect the value of the financial instruments and other contracts which are referenced to them. Manipulating the value of a benchmark therefore results in a direct transfer of money from one of the contracting parties to the other. The risk of manipulation observed in many benchmarks is driven primarily by the discretion submitters have when selecting data for submission to the calculator or when assessing their submissions, and the discretion that administrators exercise when calculating the benchmark. Some benchmarks rely exclusively on assessments from contributors, with little possibility for ex post verification against real data. The provider may not be able to verify whether these submissions are representative and applies discretion in assessing or calculating the benchmark on the basis of this data.
Moreover, contributors may choose to submit only a subset of their data, which hurts the accuracy of the benchmark. Contributors may also be reluctant to provide complete or accurate data to benchmark administrators if this conflicts with other interests, such as disclosing sensitive or confidential commercial data or information which acts as a signal to other market participants. Conflicts of interest exist in particular where the contributors or administrators are also the users of the benchmark. Contributors may submit insincere estimates, or false or partial data which do not necessarily reflect true market conditions, when they are able to profit by changing the benchmark.
Where an index is published, its use may become so widespread that it becomes critical, or at the least any failure will have significant impact on markets and investors. Finally, most authorities do not have supervision and enforcement powers over the setting processes and actors, which prevent the effective oversight of benchmarks and enforcement of compliance with minimum standards in their provision.
2.2. Use of benchmarks which are not robust, reliable or fit for purpose
Where a benchmark is not robust and is subject to the risk of manipulation, its use may harm investors or other users. More sophisticated users, such as banks and other wholesale market participants, may have a good understanding of the risks posed by the benchmarks’ lack of robustness and absorb this risk, or be able to take appropriate mitigating measures. However retail consumers may not be fully informed of the nature of the benchmark to which a financial contract they enter into is referenced and may not have any choice about the benchmark used.
The problem of the use of benchmarks which are not robust, reliable or fit for purpose is mainly driven by the lack of transparency about the purpose of benchmarks and the lack of suitability of benchmarks for their intended purpose, particularly for retail financial contracts.
Benchmarks measure a particular market or economic reality. However, in many cases, there is a lack of transparency about what exactly a benchmark is intended to measure, in what circumstances it provides a reliable measure and about other risks associated with its use. If this transparency is lacking, users may not be able to make appropriate decisions about which benchmark to use. When this occurs, economic decisions will be based on distorted values, leading to a less than optimal allocation of assets and potential losses for investors and consumers.
Often benchmarks are used to reference consumer contracts without sufficient or appropriate assessments of their suitability for this purpose. Even benchmarks which adequately measure the economic reality which they are intended to represent may have a harmful impact when used for other purposes. This problem is driven by a lack of understanding on the part of those entering into contracts referenced to them, especially when they are retail investors or consumers. Often markets use benchmarks not based on their suitability, but simply because network effects make a particular benchmark the established unit of measure. Similarly, unequal bargaining power or the use of standard terms and conditions mean that choice of an inappropriate benchmark may be imposed on one of the counterparties. There may also be significant conflicts of interest in strategy/proprietary indices which are used or produced by fund managers who have direct interest in the performance of these funds.
3. Subsidiarity and proportionality
While some benchmarks are national, the benchmark industry as a whole is international in both production and use. For benchmarks that are widely used or produced across several Member States national action may lead to fragmentation of the internal market and could facilitate regulatory arbitrage, as benchmark production can be easily moved to other Member States. By contrast, an EU initiative would help enhance the single market by creating a common framework for the use of reliable and suitable benchmarks across different Member States. EU action is therefore in line with the principle of subsidiarity.
The principle of proportionality requires that any intervention is targeted and does not go beyond what is necessary to achieve the objectives. At the identification of alternative options, as well as throughout the analysis and comparison of options and their scope, the proportionality principle has been guiding the process. This has been achieved in two ways: firstly, by targeting only those benchmarks that may have a direct and certain economic impact if they are manipulated and secondly, by identifying measures which by their nature do not involve issues of proportionality or, where measures might impose a disproportionate burden, enabling those measures to be calibrated in a proportionate fashion.
The scope of the proposal has been determined by reference to the problems outlined above to ensure that the problems are effectively, efficiently and proportionately addressed. The proposal’s scope is determined in a two-step approach: first by reference to which benchmarks (“targeted benchmarks”) are covered and secondly, which activities in relation to these targeted benchmarks should be covered.
First, all published indices which are used to reference the price of a financial instrument or contract or measure the performance of an investment fund are covered since these are the benchmarks whose manipulation will cause financial loss and which are most susceptible to the problems outlined.
Second, all benchmark administrators producing these benchmarks based in the Union are within scope, as administrators are subject to all the problems outlined above. Central banks are exempt from the scope as they already respect a high level of integrity and are subject to control by public authorities. Contributors based in the Union are also subject to these problems but according to the analysis it is only proportionate to target contributors that are already subject to EU financial regulation. The main reason is that the compliance burden would mean that currently unregulated contributors could potentially cease to contribute and in any event could not be effectively supervised by financial regulators.
The table below provides an overview of the general, specific and operational objectives of this initiative in light of the analysis of the risks and problems above:
General objectives || Specific objectives || Operational objectives || Measurable objectives
1) Restore confidence in financial markets 2) Enhance market integrity and fairness; ensure that financial markets serve the real economy 3) Ensure consumer and investor protection || Ensure benchmarks’ integrity: 1. Reduce the risk of benchmark manipulation 2) Ensure the appropriate use of robust and reliable benchmarks || 1) Limit incentives and opportunities for manipulation 2) Minimise discretion - ensure benchmarks are based on sufficient and representative data 3) Ensure robust governance and controls address risks 4) Enhance transparency and the use of suitable benchmarks 5) Ensure effective oversight || 1) Reduce the number of benchmarks vulnerable to manipulation 2) Increase the number of benchmarks based on sufficient and representative data 3) Reduce the number of findings of inappropriate governance and controls 4) Increase the number of statements on benchmarks purpose and assessments of suitability for retail use 5) Reduce the number of cases of benchmark manipulation
6. policy options and analysis of the impacts
The different policy options were tested against the criteria of their effectiveness and efficiency in achieving the six operational objectives.
6.1. Limit incentives and opportunities for manipulation:
1 No action
2 Manage and disclose conflicts of interest
3 Structural separation
Option 2 is the preferred option: structural separation is effective in limiting the opportunities for manipulation, but it would be disproportionate in many cases, as it could discourage benchmark provision and lead to the discontinuation of some benchmarks. Managing and disclosing conflicts of interest creates the incentives for those performing calculations or submitting information to benchmarks to do so in an objective manner. Thus, it encourages accurate submission of data and honesty in the data validation process.
6.2. Minimise discretion; ensure benchmarks are based on sufficient & representative data:
1 No action
2 Require the use of transaction data if available and reliable, otherwise well founded and verifiable discretion
3 Mandatory use of transaction data only if necessary
4 Mandate contributions for critical benchmarks
Options 2 and 4 are the preferred options: a rule requiring that representative transaction data take precedence, and that the use of other data needs to be verifiable, provides an instrument for ensuring that benchmark rates and contributions are based on sufficient and representative data and that discretion is justified, well founded and properly exercised. Thus, it enhances the reliability of benchmarks and transparency of the process. The power to mandate contributions to critical benchmarks ensures the continuity of benchmarks which are of critical importance when there is the risk of contributors ceasing their contributions. It also ensures that they are based on sufficient samples of market data, particularly in times of market stress when contributors may be reluctant to submit data or estimates for benchmark calculation.
6.3. Ensure robust governance and controls address risks:
1 No action
2 Supervisory authorities to issue comply or explain guidelines
3 Mandate adequate management systems and effective controls
Option 3 is the preferred option: given the inherent conflicts of interest in making contributions to and calculating benchmarks, comply or explain principles are not likely to achieve the desired results. Mandating adequate management systems and effective controls provides a tool for enforcing robust governance and controls necessary to address the shortcomings identified in benchmark provision and contribution activities.
6.4. Enhance transparency and ensure the use of robust and reliable benchmarks:
1 No action
2 Require transparency on methodology, underlying data, process and purpose, whilst allowing for delayed or partial transparency of underlying data when justified
3 Assessment of suitability of benchmarks’ use for retail contracts
4 Mandatory notification of benchmarks use
Options 2 and 3 are the preferred options: transparency on data and methodology would allow both the regulators and the public to evaluate whether the benchmark is accurate and reliable. With access to both the data and the methodology, investors and regulators would be able to replicate or back test the benchmark in order to assess its accuracy. Full transparency about what the benchmark measures, how it should be used and its shortcomings would enable regulators and the public to be fully informed about the economic reality a benchmark is intended to measure and of any shortcomings it may have in tracking this. Delayed publication or partial publication would be allowed if full and contemporaneous publication would result in serious adverse consequences for the contributors or adversely affect the reliability or integrity of the benchmark. Publication would only be permitted to be delayed to the extent it significantly diminished these consequences. Where a regulated entity such as a bank intends to enter into a financial contract with a consumer, where the payments are referenced by a benchmark, it should assess the suitability of the benchmark for this use and warn the consumer if it is unsuitable.
6.5. Ensure effective supervision of benchmarks:
1 No action
2 Private benchmark provision, independent private oversight
3 Private benchmark provision, public supervision and enforcement
4 Public provision of critical benchmarks
Option 3 is the preferred option: public provision of benchmarks may hinder innovation and may also be subject to conflicts of interest. Private oversight may also be subject to conflicts of interest. Making the provision of benchmarks a regulated activity ensures the supervision of benchmark administrators to the highest possible level. It would thus provide authorities with a tool for enforcement of compliance and accountability of administrators and contributors. It would ensure the independence of the supervisor as it would be a public authority and allow for market choice, innovation and competitiveness as benchmark provision would remain in private hands.
Supervision of administrators would be carried out by the national supervisory authorities under the coordination of ESMA. For critical benchmarks, supervision by a college of national supervisors would be mandatory. ESMA would participate is the colleges of supervisors for critical benchmarks and exert binding mediation when necessary. Supervision of contributors would be exercised by their current supervisors.
7. Impacts of the preferred options package
The chart below administrators a summary of the retained policy options
Operational objective || Preferred policy options
Limit incentives and opportunities for manipulation || Manage and disclose conflicts of interest
Sufficient data and minimised discretion || Transaction data if available and reliable, otherwise well founded and verifiable discretion
Mandate contributions for critical benchmarks only if necessary
Ensure robust governance and controls address risk || Mandate adequate management systems and effective controls
Enhance transparency and ensure the use of benchmarks which are robust, reliable and fit for purpose || Require transparency on methodology, underlying data, process and purpose, whilst allowing for delayed or partial transparency of underlying data when justified
Assessment of suitability of benchmarks’ use for retail contracts
Ensure effective oversight || Private benchmark provision, public supervision and enforcement
The main goal of this initiative is to enhance the robustness and integrity of European benchmarks by reducing the risk of benchmark manipulation. Thus, it will contribute to enhancing market integrity and fairness and to restoring confidence in financial markets. This initiative also aims to ensure the appropriate use of robust and reliable benchmarks, which will enhance consumer and investor protection and assure that financial markets serve the real economy.
8. Monitoring and evaluation
The evaluation of the consequences of the application of the legislative measure could take place three years after the entry into force of the legislative measure, in the context of a report to the Council and the Parliament on the effectiveness of the legislative initiative and appropriateness of the sectoral approach.