FCA recommends transition away from LIBOR

Summary

On 27 July 2017, the UK Financial Conduct Authority (the FCA) published a speech by Andrew Bailey, its Chief Executive, on the future and sustainability of LIBOR (the London Interbank Offered Rate). 

In summary, the FCA has said that:

• it does not consider that LIBOR is sustainable;
• the market should work to transition to alternative rates by the end of 2021; and
• firms should not rely on LIBOR being available after the end of 2021.

Background

Since the FCA started regulating LIBOR in April 2013 significant amendments have been made to its administration. One of the key aims of the reforms has been to seek to ensure that LIBOR rates are more representative of market conditions by requiring panel banks to base their submissions to the greatest extent possible on actual transactions. However, a lack of activity in the underlying market that LIBOR measures has made this harder than expected to achieve. For example, as explained in Mr Bailey's speech, in one currency–tenor combination, for which a benchmark reference rate is produced every business day using submissions from around a dozen panel banks, the panel banks executed just fifteen transactions of potentially qualifying size in that currency and tenor in the whole of 2016 between them. Consequently, LIBOR is currently sustained by the use of “expert judgement” by the LIBOR panel banks to form many of their submissions.

The FCA considers that this raises a serious question about the sustainability of LIBOR. As Mr Bailey put it, "if an active market does not exist, how can even the best run benchmark measure it?". Furthermore, panel banks are increasingly uncomfortable with providing submissions based on judgements with so little actual borrowing activity on which to base those judgements. According to the FCA, panel banks' submissions have for some time now been driven by the FCA, which, concerned that a withdrawal of panel bank support could further weaken the representativeness of LIBOR, has been "strongly encouraging" continued submissions from panel banks in recent years.

The FCA's announcement does not definitively signal the end of LIBOR (see "What is the alternative?" below), but it does strongly recommend a transition away from LIBOR to alternative benchmarks that are more representative. Current panel banks have agreed to continue their LIBOR submissions until the end of 2021, the idea being that by this date the market will have transitioned away from LIBOR such that these submissions will no longer be necessary. The proposed transitional timeline reflects (i) the FCA's acknowledgement that any transition away from LIBOR must be planned and gradual to avoid the major disruption that would be caused by its sudden discontinuation; and (ii) the FCA's belief that the market will only begin preparation for transition in earnest if there is a specific date to aim for.

What is the alternative?

Alternative reference rates have been the subject of much discussion in recent months, with SONIA (the Sterling Over-Night Index Average) and a new broad Treasuries repo rate being designated as the preferred alternative benchmarks by the Bank of England's Risk Free Rate Working Group and the United States' ARRC (Alternative Reference Rates Committee), respectively¹. However, there are some fundamental differences between both of these rates and LIBOR which are already causing concern in the market. Of course, as Mr Bailey points out, IBA (ICE Benchmark Administration Ltd.), LIBOR's administrator, may decide to continue to administer LIBOR after 1 January 2022, if it is able to do so. However, given the shortcomings highlighted by the FCA, it seems unlikely that the continuation of LIBOR will be feasible in the long-term. It would appear that these alternative rates, which were originally intended to help the derivatives market wean itself off LIBOR, may now be considered for a far wider purpose. Ultimately though, their success will depend on the uptake of end-users to create sufficient liquidity for them to be considered viable options.

One key question that Mr Bailey raises in his speech is this: is the better approach (i) to amend contracts to reference an alternative rate, or (ii) to use a protocol to amend the definition of LIBOR and replace the current methodology with alternative reference rates? Whatever Mr Bailey's view on this, the FCA has made it clear that it is now up to market participants to decide, and furthermore to take such action as is necessary to ensure that the market is in a position by the end of 2021 to manage without LIBOR, should it be discontinued. Some sectors, such as the OTC derivatives industry, are ahead of the curve, having contemplated the possible discontinuation of key IBOR rates and the potential effects thereof for some time already. Others may find themselves less well positioned.

ISDA position

Prior to the last week's speech, ISDA had already established working groups to consider how the permanent discontinuation of an IBOR might affect the derivatives industry, focusing predominantly on (i) how the impact of any such discontinuation could be mitigated, and (ii) what methodology should be used to determine rates and prices under derivatives documentation should a referenced IBOR be discontinued.

Most derivatives contracts incorporate by reference one or more sets of standardised ISDA definitions, each of which is specific to a particular type of product. For example, the documentation relating to an interest rate swap will incorporate by reference the 2006 ISDA Definitions (the 2006 Definitions). Although described as "definitions", these documents, which form part of the contracts into which they are incorporated, contain important operational provisions, including provisions designed to deal with the scenario where a particular rate is not published in the usual way.

Current fallbacks

Under the 2006 Definitions, the fallback position where a LIBOR rate is not available is that the applicable rate is determined, by the party acting as the calculation agent, on the basis of comparable rates offered by four major banks in the London market. If it is not possible to obtain rates from four such banks, the calculation agent has the discretion to select other major banks in London from which to obtain rates. A similar fallback mechanism applies in respect of foreign exchange and currency trades, as set out in ISDA's 1998 FX and Currency Option Definitions.

Recent developments

Following a request from the Financial Stability Board (the FSB) in July 2016, ISDA has been working towards developing more robust fallbacks to afford counterparties increased certainty in what ISDA had previously described as "the unlikely event that a key IBOR is discontinued". The working group's focus has hitherto been on USD, GBP and Yen and they had recently concluded that the preferred alternative rates selected by the UK, US and Japanese working groups would be the most logical fallback rates in the event of a discontinuation. No specific timeline has yet been implemented to effect the necessary documentation changes, but, now that this "unlikely event" is looking more likely, developments in this regard are likely to speed up and we expect to see increased activity from ISDA on this front in the coming months.

Any changes to fallbacks would require amendments to be made to a range of ISDA documentation, which would be effected using ISDA's usual consultation and amendment process. ISDA's view is that, ideally, all market participants would transition to the same fallbacks and successor rate(s) at approximately the same time, to avoid creating basis and associated risk. For similar reasons, the new regime should apply equally to new and legacy trades. Traditionally, ISDA has used protocols to assist its members in making multilateral amendments to its existing contracts, and it has been suggested that this approach could also be used here. However, adherence to ISDA protocols is voluntary so, absent any regulatory compulsion to adhere or to effect equivalent amendments bi-laterally, there is a possibility that some market participants would choose other fallback mechanisms and/or alternative rates.

Repo and stock lending markets

GMRA

The 2000 version of the global master repurchase agreement (GMRA) contains a definition of LIBOR, which is used to calculate interest in certain scenarios. The definition refers to a specific (former) Telerate screen and the fallback, such as it is, is to "such other page as may replace [that page]". The definition has been removed from the 2011 version of the GMRA and the function of LIBOR as used in the 2000 version has been replaced with the concept of an "Applicable Rate", which affords the non-defaulting party more flexibility in selecting the rate to be used for its calculations.

Parties frequently reference LIBOR, which is a defined term under the 2000 and 1995 versions of the GMRA, in transactions when establishing what is known as the "Pricing Rate" – a rate which is used to determine the repurchase price of the securities which are the subject of a particular transaction. Unless the parties have chosen to use established ISDA terminology or definitions (see "Current fallbacks" above) when setting out this Pricing Rate, often no fallback mechanism is specified. As a result, if LIBOR is to be discontinued entirely, these contracts may need to be updated and amended to stipulate how LIBOR-based calculations are to be managed thereafter. However, given that repurchase transactions are typically relatively short-term transactions, and that the market will most likely receive ample notice of any impending discontinuation, it seems unlikely that parties to repurchase transactions will find themselves with large numbers of legacy trades referencing LIBOR once it has disappeared. However, there will inevitably be a number of longer-term transactions which will require review and, possibly, amendment.

GMSLA

Similar issues to those described above could arise in respect of stock lending transactions documented under the global master securities lending agreement (GMSLA) if LIBOR were to be discontinued. The published agreements do not contain a standard definition of LIBOR, although the concept may be incorporated in the definitions of rates payable on collateral balances. Again, as stock lending transactions are typically short-term, it seems unlikely that counterparties would find themselves with large numbers of legacy trades referencing LIBOR after its demise, but any longer-dated loans would need to be considered.

Debt Capital Markets

In the international market for plain vanilla corporate or government debt securities the methods of determining the interest rate on a floating rate note for any interest period can be divided into two broad categories: ISDA Determination and Screen Rate Determination.

Where ISDA Determination is used, the rate of interest for any particular interest period will be the sum of the specified margin and the relevant ISDA Rate; that is, a rate equal to the Floating Rate (as defined in the 2006 Definitions) that would be determined by the calculation agent under an interest rate swap transaction if the calculation agent were acting as calculation agent for that interest rate swap transaction under the terms of an agreement incorporating the ISDA definitions. In this situation, changes to, or the disappearance of, LIBOR would be dealt with in the same way as any other contract referencing the 2006 Definitions (see "Current fallbacks" above).

Where Screen Rate Determination is used, the rate of interest for any particular interest period will be the sum of the specified margin and the arithmetic mean of the reference rates which appear on the relevant screen page as of the relevant time. For these purposes the relevant screen page is usually defined as the page, section or other part of a particular information service (such as Reuters) specified in the terms and conditions (or the relevant final terms, in the case of an issue under a programme) or such other page, section or other part as may replace it, as nominated by the provider or sponsor for the purpose of displaying rates or prices comparable to the reference rate. This sort of provision would appear to provide much scope for a substitute rate for LIBOR to be displayed in such a way as to fall within such a definition and therefore provide continuity over the transitional period should LIBOR be replaced by one or more alternative rates and cease to be published.

Loan Markets

Loan markets participants, be they traditional bank lenders or the direct lending funds which have seen an explosion in activity since the Wheatley review of LIBOR in 2012, have shown no signs since 2012 of detaching their interest returns from LIBOR. For every LIBOR flaw, there is a sticking plaster – never an amputation. Two critical questions now arise: what alternatives to LIBOR will the loan markets envisage when forced to do so? And how will these alternatives be brought into effect in legacy documents?

Loan Market Association (LMA) standard documentation currently provides a menu of alternatives if LIBOR (as administered by IBA) is unavailable. Certain of these alternatives deal with shortcomings in the benchmark itself. For example, where there is no rate available for a given period, LMA documents provide for interpolation between tenors for which a rate is available. The alternatives which might deal with the complete discontinuation of LIBOR are effectively two-fold: firstly, the benchmark defaults to a "Reference Bank Rate"; that is, the arithmetic mean of quotations provided by four reference banks (potentially fewer). The problem is, who wants to be a reference bank? Secondly, if no Reference Bank Rate is available (there is no compulsion to quote), then each lender may notify its own "Cost of Funds" i.e. the cost to that lender of funding its participation from whatever source it may reasonable select. Those notified rates may be passed on to the borrower directly – or documentation may provide for a weighted average rate to form the new benchmark "LIBOR".

Prior to the discrediting of LIBOR, the Cost of Funds "solution" was intended as a final backstop – to be implemented in circumstances of market disruption when there was no LIBOR screen rate available. It was seen as a temporary sticking plaster and certainly not a long-term solution. Indeed, the LMA anticipates that borrowers may require the right to revoke a utilisation request where it transpires that pricing will be on a cost of funds basis. Cost of Funds raises issues for lenders too, since a lender's cost of funds is a highly sensitive matter. Whilst there are provisions in LMA documentation which protect the general confidentiality of funding rates, this of course does not extend to the borrower – who has a legitimate interest in knowing which lenders are, in effect, offering cheaper or more expensive debt. Widespread use of Cost of Funds as a solution would inevitably create downward pressure on that pricing element. With margins already at historic lows relative to probable risk, it is not an attractive option for the loan markets.

What other options are there? The use of SONIA (for sterling) or other overnight rates is a possibility – but it introduces a disconnect between the tenor of the benchmark and that of the funding provided. Is that a worse disconnect than that presented by LIBOR in its current incarnation, which is something of a "guesstimate"? Maybe not.

Four years (taking us to the end of 2021) is quite a long time in the loan markets. Even allowing a number of months for a LIBOR Plan B to gain some traction, many (probably most) LIBOR-based transactions will expire or be amended (as part of the natural loan cycle) in that period, so there will be an opportunity to update provisions under a typical majority Lender consent regime. Lenders (especially the mainstream bank lenders) prefer to move en masse though, so we expect the LMA to be engaging with market participants to try to reach a broad consensus. If that process is too slow and unwieldy, major financial borrowers may take the lead and decide for themselves what works (for them) – and then impose that solution on a highly liquid market. In effect, early LIBOR defectors may forge the path along which the rest will follow.

EU Benchmark Regulation

The EU regulation on benchmarks²  (the Benchmark Regulation) entered into force in June 2016 and will become fully applicable in the EU on 1 January 2018³, subject to certain transitional provisions. When fully applicable it will impose wide-ranging obligations on the administrators of benchmarks which are used in the EU and those entities which contribute data to them. It will also restrict the use of non-compliant benchmarks in the EU, including the referencing of such benchmarks in securities, derivatives contracts and consumer loans which are in scope of the Benchmark Regulation. Benchmarks administered by non-EU administrators will only be able to be used in the EU if (i) an "equivalence" decision has been made in respect of the country in question, (ii) the administrator has been "recognised" in accordance with the Benchmark Regulation, or (iii) the benchmark has been "endorsed" by an EU entity in accordance with the Benchmark Regulation (for more on the Benchmark Regulation, see our most recent briefing here).

Although the Benchmark Regulation will apply in full in under six months' time, there are still many unanswered questions troubling market participants. For example, one of the requirements is that supervised EU entities using a benchmark in the EU maintain "robust written plans" detailing the steps that they would take if a benchmark that they were using were to be discontinued. This includes, where feasible and appropriate, nominating one or more fallback benchmark(s) to be referenced in such an event.

These plans are also required to be reflected in contractual relationships with clients. It has recently been unofficially confirmed by the European Securities and Markets Authority (ESMA) that this requirement only applies to new contracts (i.e. those entered into on or after 1 January 2018), which is welcome news for market participants which had been contemplating the enormity of the potential re-papering exercise ahead of them. However, other practical issues remain; for example, how will benchmark users be able to establish coherent and lasting back-up plans in January 2018 when they do not know (i) what rates are going to be established in the future and how they might develop (for example, the Bank of England is in the process of reforming SONIA, a process which is unlikely to be completed until April 2018), (ii) whether any such rates will be adopted by the market to such a degree that they constitute viable fallbacks, or (iii) how market standard documents are going to evolve in the coming months and years (see "Recent developments" above)?

Finally, we cannot discuss an EU regulation without mentioning Brexit. Post-Brexit, the UK will be a "third country", meaning that for any UK-produced benchmark to be used in the EU it will need to fall within one of the above regimes (equivalence, recognition or endorsement). LIBOR would almost certainly have benefitted from either endorsement or recognition as it is so widespread and so heavily used already in the EU market. But would another, newer and less widely used replacement rate? It is impossible to say - and if the answer were no, any such benchmark would be excluded from use in the EU.

Conclusion

It would appear from the FCA's speech that LIBOR is now living on borrowed time. Whilst no hard deadline has been set for its actual discontinuation, we should not rely on LIBOR being available long-term. So, what can firms do at this stage to prepare themselves? Although we are unlikely to have answers to the questions raised by the FCA's speech for some time, there are practical steps which can be taken now. Due diligence will need to be undertaken to determine (i) which contracts reference LIBOR, (ii) where fallbacks are specified, whether they are sufficient, and (iii) where fallbacks are not specified, how this can be managed. Market participants will also need to ensure that they have sufficient resources in place to follow developments and implement any proposed solutions in a timely fashion. Similarly, firms can assist in the transition process by engaging with ISDA and other industry bodies to help them to find a practical, market-wide solution - this appears at present to be the best way forward.