Firstly let me thank the Chartered Institute for inviting me to speak to this essay question. Do I need to worry about benchmark regulation? Actually, I’d like to begin my answer with a different question, and a reminder of what has gone before. Did we need to worry about a world without benchmark regulation?
Sadly, I think that question was answered for us. And the answer was that we did not worry enough.
The misconduct which occurred around benchmarks requires little introduction. We can let the traders speak for themselves:
That was the largely unregulated benchmark space of pre-2013. The aim of benchmark regulation is, of course, a world where we have fewer not more worries.
There will, however, be some work required to get to that less worrisome world.
The problems we should have been worrying about were not confined to the ‘ethical drift’ seen in the behaviour of those traders quoted above. Of course there were important governance and controls issues too. Those problems, and the effort to put them right, our work to see that the right cultures are established and maintained – work which, for the avoidance of doubt, is very much ongoing – have grabbed the headlines. But there were also weaknesses in the design of benchmarks. As we saw with Libor reference rates, for longer tenors in particular, these reference rates had failed to evolve with changing markets. In this case, quote-based-reference rates were no longer adequately supported by transactions in the London interbank market at 11 o’clock.
In 2012, as a result of the scandals, Martin Wheatley was asked to conduct a review into Libor. The resulting report, the ‘Wheatley Review’, recommended criminal sanctions for benchmark manipulation, and a program of reform, rather than replacement, for Libor including its transfer to a new administrator, improvements in the methodology for calculating the benchmark, and that administration of and submission to Libor be made regulated activities.
The ‘Wheatley Review’ recommended criminal sanctions for benchmark manipulation, and a program of reform
Libor’s administrator and its 20 panel banks became regulated for benchmark activities on 1 April 2013. And in April 2015 the list of regulated benchmarks in the UK was enlarged to include seven other major benchmarks. These are the LBMA Gold Price, and LBMA Silver Price, the ICE Swap Rate, SONIA, RONIA, the WM/Reuters London 4pm Closing Spot Rate and the ICE Brent Index.
The administrators of these benchmarks are subject to FCA rules which require them to put in place appropriate governance arrangements, including the formation of an oversight committee, to have effective controls to ensure the integrity of input data and to keep adequate records. Benchmark submitters are subject to similar requirements to ensure that they manage any conflicts of interest and are subject to an annual audit. Both administrators and submitters need to have a named individual responsible for compliance with these rules.
There have been significant and concrete changes in these benchmarks as a result. Auctions which for decades have been behind closed doors are now on transparent electronic platforms – for example the Gold and Silver benchmarks. Submission-based benchmarks have transitioned to more transaction-based methodologies, for example the ICE Swap Rate.
While we have already taken decisive action in the UK, benchmarks are of course cross-border in nature, and international work is underway too. IOSCO’s global Principles for Financial Benchmarks, published in July 2013, have set a common cross-jurisdictional framework for good market practice.
These principles have been widely adopted as industry standards. They impress upon administrators the need for strong governance, robust benchmark design, transparent methodologies and clear accountability.
The Financial Stability Board has also published a number of reports focusing on specific, systemically important benchmarks. Its report, Reforming Major Interest Rate Benchmarks, published in July 2014, recommended that key interest rate benchmarks, in particular Euribor, Libor and Tibor, evolve their rates to be better anchored in transaction data. The details of these changes are currently being consulted on by the administrators and we expect to see some of them implemented in the year ahead.
The report also recommended the creation of alternative ‘risk free’ rates to improve choice for market participants and reduce dependence on Euribor and Libor. In the UK, work is on-going involving members of the industry, the FCA and the Bank of England to implement this recommendation for the sterling market. Parallel work streams are in place in the euro area, the United States, Switzerland and Japan.
The FSB also made important recommendations in its report on Foreign Exchange Benchmarks, which called for changes both in the methodology and oversight of key foreign exchange benchmarks, but also in how they are used by market participants.
And, of course, there have been important European initiatives: the joint EBA-ESMA Principles for Benchmark Setting Processes in the EU published in June 2013, and the forthcoming EU Benchmarks Regulation, which I will turn to in a moment.
The aim is clear: high standards of design and governance to improve the integrity of benchmarks, and therefore to sustain confidence in the financial markets they support.
Let us turn to the substance of that new EU Benchmarks Regulation.
Proposed in September 2013 by the European Commission, the EU Benchmarks Regulation is now in its final stages.
The first important question is what will be a “benchmark”. The answer to this does not hinge on what the benchmark references, but instead on how it is used. The regulation will capture those benchmarks which are referenced in (1) financial instruments traded on trading venues, (2) mortgage and consumer credit contracts or (3) used to measure the performance of investment funds. This will capture the UK specified benchmarks but may also include, for example, the FTSE 100, Dax and CAC 40, the Baltic Dry indices, catastrophe and weather benchmarks, and commodity benchmarks ranging from precious metals to food stuffs.
Benchmarks produced in third countries but used in the EU will need to meet similar standards.
There are a number of important exemptions. Benchmarks which reference a price of a single financial instrument (such as the price of a specific share) will not be in scope, nor will benchmarks provided by central banks (such as the Base Rate) or other authorities for public policy purposes (including standard measures of inflation). Standard variable rates offered by credit institutions will not be considered benchmarks when used in mortgage or consumer credit contracts.
Where benchmarks are combined together to form a new rate, this will be considered ‘use’ of these benchmarks rather than administration of a new benchmark. This is of particular significance to asset managers.
Note also that this definition will not capture all benchmarks. For example, a benchmark used to support only instruments which are not traded on trading venues, would not be in scope.
The regulation will supersede the UK’s existing rules – covered in Chapter 8 of the FCA’s market conduct section of the FCA’s Handbook – and affect a wider set of firms than our current rules. These can be divided into three main groups: benchmark administrators, contributors and users.
The regulation will require the authorisation or registration of benchmark administrators, who will then be subject to supervision against the requirements in the regulation. These requirements, like the IOSCO principles, cover governance and accountability as well as the design and methodology of any benchmarks provided.
Benchmark contributors who already conduct regulated financial activities and provide data which are not readily or publicly available will need to meet additional requirements to ensure the integrity of their submissions. Their governance and systems and controls will be subject to supervision by national competent authorities.
Users who are already supervised entities – including credit institutions, investment firms, UCITS and AIFMs – will no longer be allowed to use a benchmark unless it is provided by an authorised or registered administrator in the EU or, in the case of third countries, the third country administrator has been recognised or the benchmark has been endorsed.
This worry about whether reference prices and benchmarks produced in third countries will no longer be accessible to EU firms because of equivalence requirements that these jurisdictions seemed unlikely to meet, has been a key concern. “Will I still be able to use these benchmarks – say the S&P 500 – as a benchmark for my fund’s performance?” “Can I still buy or sell options on the S&P 500?”, and similar, have been key questions.
The answer to these questions is that there are mechanisms available for all benchmarks to be admitted for use in the EU but these do require action by their administrators to either become ‘recognised’ in the EU (by a relevant national competent authority) or to have their benchmarks ‘endorsed’ by an EU supervised entity who would then become responsible for ensuring that benchmark was provided in a manner consistent with the regulation. We would expect all the major non-EU benchmark administrators to utilise one of these routes. It is possible, however, that some benchmarks will not pass this test, in which case it will not be possible to reference these benchmarks.
Supervised users will also need to ensure that they have in place written plans in case the benchmarks they are using cease to be produced. These plans would include designating alternative fall-back rates, if and where appropriate.
As with other EU regulations, we urge firms not to wait until the Level 2 process is finished before taking steps to focus on the risks the regulation is seeking to address and how you should respond. We also urge firms not to focus only on what is now set to be in scope of the regulation. Indices that are outside the perimeter of the new regulation may also present risks for which controls could be important.
Whilst the regulatory approach taken by Europe may be more comprehensive and prescriptive than some other regimes, the aims and expectations are essentially those set out in the IOSCO Principles for Financial Benchmarks.
It is important that you don’t think this regulation is only a matter for benchmark administrators, or submitters. Users will be affected too. Where administrators fail to become authorised or registered in the EU, their benchmarks will no longer be available for use.
Benchmark users are the party best placed to identify and manage this risk, to minimise disruption to their own business. There are two important steps involved.
It is important that you don’t think this regulation is only a matter for benchmark administrators, or submitters. Users will be affected too.
Firstly, users should make sure they are aware of which benchmarks they use, and seek assurances from the administrators of these benchmarks that they are aware of the regulation and have plans to comply with it. This is particularly important where the benchmarks are provided by smaller administrators, especially in third countries.
Secondly, both in anticipation of the future requirements and for clear reasons of prudence, users should start thinking now about potential alternative benchmarks in case any benchmark they currently use becomes unavailable. Contingency plans such as robust contractual fall back clauses are necessary in case a benchmark cannot be used, or is no longer available.
This issue of benchmarks no longer being available is one that existed before the Benchmark Regulation. It is one revealed by the Libor experience. One of the most important lessons from that episode, which we should not lose sight of, is how the benchmarks had become detached from the underlying markets they were developed to measure. Some tenors and some currencies have been discontinued. In other cases, the reference prices may continue to exist but may no longer be the best way of capturing the economic risks that users wish to hedge.
Even where Libor is the best hedge, we must be conscious that some very large derivatives positions are in effect based on some fairly slender foundations in terms of the underlying activity measured by the benchmark. The Libor-based interest rate derivatives market, for instance, is by some measures more than one thousand times bigger than the unsecured interbank money market which Libor represents. This informs the importance of making those foundations as strong as we can, but also the importance of being prepared for if and when those benchmarks can no longer satisfy the intended purpose.
Note that this is not a problem that is inherently confined to interest rate benchmarks. For example, if you are running an index to measure shipping costs, perhaps based on a standard size or type of ship, how do you evolve that benchmark to take account of bigger and/or faster vessels? If you run an oil index, what happens when the supply from a particular oil field is at or near exhaustion?
Regulation and other international workstreams in relation to benchmarks will try and ensure some minimum standards for benchmarks, make them less easily manipulated, more transparent, and more robust. Some benchmarks may not prove strong enough to meet those new standards, and could fall away. But it is users themselves who must take prime responsibility for ensuring benchmarks they use are appropriate, and that they have plans for the event that these are no longer available.
Users have a vital role to play to ensure that the implementation of the regulation is smooth and succeeds in delivering high standards
Alongside regulators and administrators, users have a vital role to play to ensure that the implementation of the regulation is smooth and succeeds in delivering high standards with minimal disruption.
Some administrators and benchmark users are already ahead of the game vis-a-vis the EU Regulation, for example benchmark administrators who have already begun to raise their standards under IOSCO Principles ahead of the EU Benchmark Regulation, and those users who have already carefully thought through their benchmark use, and worked sensibly to move away from benchmarks that could no longer be relied upon.
For other administrators and users, some or much of this will be a new exercise, and represent a raising of the bar. Because of this, and because in implementing this regulation, the FCA will be looking to support administrators, contributors and users of benchmarks in meeting the new standards, the FCA team will be reaching out to market participants to provide information that will help them, and is equivalently keen to learn from market participants’ concerns. This event is one small contribution to that, but I would very much encourage all who have a stake in this market to add their names to the dedicated mailing list which we will be using to update on the regulation.
Copyright © 2016 FCA. All Rights Reserved.