Work continues on the transition to risk-free rates, but progress is slow—FCA publishes timetable and milestones.
LMA issues guidelines, but changes may not be readily accepted by market participants.
A “synthetic” or “zombie” LIBOR will likely continue to be published in order to deal with legacy transactions.

On July 27, 2017, Andrew Bailey, Chief Executive of the UK Financial Conduct Authority (FCA) announced that because of insufficient trading in the underlying markets, LIBOR (the London Interbank Offered Rate) will no longer be supported by the FCA after 2021.

LIBOR has been under pressure ever since the financial crisis of 2007-2008 when the confidence in banks failed and banks effectively stopped lending to one another. Since then, a number of rigging scandals have come to light (and litigation remains ongoing), and the spotlight has centred on the failings of LIBOR. Political pressure has also come to bear on the interest rate, and those financial institutions that are required to provide a rate have become ever more reluctant to do so.

LIBOR has been subject to review and scrutiny long before the FCA 2017 announcement however, and there has been a call to reform the rate for a number of years so that it more accurately reflects the true rate of interest. In 2012 for example, the Wheatley Review called for reform of the rate and for implementation of a strict process to verify rate submissions with transaction data.

There has been ongoing reform of LIBOR since the Wheatley Review made its recommendations. Most significantly, ICE Benchmark Administration Ltd (IBA) took over the administration of LIBOR in February 2014 and has undertaken a number of reforms since that time. Most recently, in April 2018, the IBA published its latest set of reforms in its publication “ICE LIBOR Evolution” in which it set out the Waterfall Methodology for the determination of LIBOR, whereby LIBOR would be determined by panel banks which are required to submit their LIBOR submissions in wholesale, unsecured funding transactions to the extent possible. To the extent that a panel bank has insufficient eligible transactions (being Level 1 of the Waterfall), it will seek to make a submission based on transactions derived data (being Level 2 of the Waterfall); failing this panel banks are required to submit a rate at which the panel bank can fund itself at 11a.m. London time with reference to the unsecured, wholesale funding market (i.e., basing its rate on transactions, related market instruments, broker quotes and other market observations).

Soon after the July 2017 statement by the FCA, participants in the market started to consider what this really meant for the various markets in which LIBOR plays a pivotal role. Some markets have moved quicker than others in developing solutions in respect of the products which they cover. Difficulty arises from the fact that all of the various financial markets—whether the derivatives market, the wholesale bond market, the loan market, or any other market—are all interconnected, making it impossible for one market to work in isolation to the others in finding a solution.

Whilst there is no clear alternative to LIBOR, focus has been on the development of and transition to risk-free rates (RFRs). This in itself has not been without difficulty in that the various RFRs are based upon different methodologies. For example, the US$ RFR, SOFR (Secured Overnight Funding Rate) is a secured rate, whereas others, such as the RFR selected in the UK, SONIA (Sterling Overnight Index Average), are not. The RFRs are themselves in a developmental stage—if they exist at all. The Euro RFR is currently being developed; the latest changes to SONIA methodology were introduced in April 2018 to make the rate more reflective of the market by capturing a broader scope of overnight unsecured deposits by including bilaterally negotiated transactions alongside brokered transactions, thereby increasing underlying transactions upon which SONIA is based to around £50 billion per day.

But SONIA and the other RFRs are not currently fit for purpose as a direct substitute for LIBOR. One of the more significant issues is that the RFRs are backward looking, being calculated at the end of the overnight period to which they relate. Clearly in order for the replacement rates to operate in the loan (and other) markets, methodology must be created so that the RFRs look forward so as to provide interest rates for terms consistent with those currently provided by LIBOR (i.e., overnight, one week, one, two, three and six months and one year). Another significant issue is that RFRs by definition are risk-free rates, meaning that the interest rates are inherently lower than LIBOR (which reflects banks’ credit risks and cost of funds). A straight swap from LIBOR to an RFR is therefore not possible unless this difference (or pricing gap) is accounted for in documentation.

The amount of work which needs to be undertaken by market participants in the period until the end of 2021 is vast and should not be underestimated. Clearly, determination of the rate to be used going forward is one thing, but documentation generally does not contemplate a change away from LIBOR and will need to be changed. To this end, a working group (and numerous sub-working groups) have been established by the FCA and progress is being made. A timeline with milestones has been developed to scope out the work being undertaken with implementation of SONIA being scheduled to take place during 2019 and transition being undertaken during 2020/2021.

The London Loan Market Association (LMA) has also been active since the 2017 FCA announcement, informing participants in the market of the challenges which the change in reference rate will bring about and sitting on various sub-working groups which have been tasked to effect the transition.

In May 2018, the LMA published its first guidance (“The Recommended Revised Form of Replacement Screen Rate Clause and User Guide”) on documentary issues which will arise in loan documentation as a result of the change and proposed slot-in drafting to be used in its template loan agreements, although this stopped well short of proposing new language for a SONIA-derived rate.

The LMA approach so far has been to suggest language which will facilitate the change to SONIA (or other benchmark rate) when the time comes so that the change can be implemented in an easier fashion. Whether this will be accepted by the loan market remains to be seen. One of the more significant obstacles which has arisen is that the proposed wording provides that the changes may be implemented by the borrower and the facility agent, the latter acting on the instructions of the majority of lenders (i.e., rather than all of the lenders). To hand over the rate of return which the lender receives, being a paramount matter for any lender, to the majority of lenders and thereby relinquishing control of the interest rate being used is a significant matter and one which may not be readily accepted.

Even if the LMA proposals are accepted by the market, only time will tell if the two-year transition period once the new benchmark rate has been settled is sufficient time to enable the financial markets to adapt to the new regime. With the thousands of loans, bonds and other products outstanding in the market it will certainly be a stretch for all to be amended in time. Certainly some financial products are easier to deal with than others—for example, ISDA is able to amend its protocol relatively easily. Loans and bonds on the other hand are not so easy to amend, and it is likely that a “synthetic” or “zombie” LIBOR will continue to be published in order to deal with the legacy transactions which remain.