In December 2019, the International Swaps and Derivatives Association (ISDA) wrote to the FSB and recommended that the FCA and the ICE Benchmark Administration (IBA) provide further clarity about the length of any “reasonable period” in which a non-representative LIBOR might be published prior to LIBOR being discontinued.
The FCA has now responded to ISDA’s recommendation and comments that market participants should not assume that any period of non-representative LIBOR based on a reduced panel bank submissions would last for more than a short period (i.e. months not years). In its letter, the regulator acknowledges that a non-representative panel bank LIBOR would be an uncomfortable position for LIBOR users as the behaviour of the rate would be difficult to predict and it would be likely to be more volatile given the smaller number of panel banks. The FCA notes that the law governing critical benchmarks, such as LIBOR, is set out in the UK Benchmarks Regulation (BMR) which will come into force at the end of the Brexit transition period (31 December 2020), assuming the UK leaves the EU as planned on 31 January. Under Article 21(1) BMR, IBA must give notice to the FCA when it intends to cease providing LIBOR and Article 21(3) gives FCA the power to compel IBA to continue production until the benchmark “can be ceased in an orderly fashion”. The FCA has confirmed that it does not expect to compel IBA to continue to produce a non-representative panel bank LIBOR beyond end-2021 for the benefit of ‘tough legacy’ contracts.
The IBA has separately written to ISDA, outlining the procedure it will follow should it suspect that certain or all LIBOR settings will become unrepresentative. The IBA notes that the period of time to effect a cessation of LIBOR will be dependent on the facts and circumstances at the time but that its preference would be that any cessation of LIBOR would be pre-announced a reasonable time in advance of the benchmark becoming potentially unrepresentative, to avoid any requirement to publish a non-representative benchmark. The IBA has confirmed that it would not be comfortable with publishing an unrepresentative benchmark and highlighted that the continuing publication of such is not contemplated by the BMR.