The past week has been important for developments in LIBOR transition, particularly for the cash markets where progress has hitherto been less advanced than other markets.
On 26 February 2020, the International Swaps and Derivatives Association (ISDA) and Securities Industry and Financial Markets Association (SIFMA) hosted a joint conference on benchmark reform in London. One of the most important speeches at this event was given on behalf of the Bank of England (BoE) by Andrew Hauser: Turbo-charging sterling LIBOR transition: why 2020 is the year for action – and what the Bank of England is doing to help.
So what is the BoE doing to accelerate the transition? It seems the answer is to offer both “carrots” and “sticks” to encourage the transition of financial products from LIBOR to SONIA. In a nutshell, the BoE has announced measures:
- Supporting the adoption of SONIA in cash products by simplifying the calculation of compounded interest rates by agents through the publication of a SONIA-linked index from July 2020;
- Acknowledging the market’s desire to go a step further by agreeing to consult the potential publication of daily “screen rates” for specific period averages of compounded SONIA which would avoid agents having to perform any calculation at all; and
- Penalising firms which continue to hold LIBOR assets by October 2020, by increasing haircuts on LIBOR linked collateral when using the BoE’s funding window, if that collateral remains LIBOR-linked.
These measures are explained further below, together with a discussion as to how these developments are likely to be received by the market, and how they might alter the profile of the litigation risk faced by financial institutions and other corporates given the impact on ease, certainty and necessity of transition.
1. BoE to publish compounded SONIA-linked index from July 2020
The BoE intends to publish a compounded SONIA-linked index from July 2020. It has said that this will complement its existing publication of the overnight SONIA rate and the index will provide a flexible tool to help market participants construct compounded SONIA rates in an easy, consistent and flexible way. The stated intention of the BoE is to support market participants to cease issuing GBP LIBOR-based cash products (maturing beyond 2021) by Q3 of 2020. This is the deadline set by the UK Risk Free Rate (RFR) Working Group in its roadmap for 2020 – see our banking litigation blog post for further discussion of the 2020 roadmap and the accompanying suite of documents published by the regulators and working groups at the beginning of this year.
The BoE has celebrated the speed with which SONIA has become the default reference rate of choice for floating-rate notes and securitisations (which are predominantly wholesale transactions), but recognises that replicating this success in the bilateral and syndicated loan market is a different challenge. It is understood that one of the challenges to widespread adoption is the practical difficulty of calculating compounded SONIA rates in a simple and consistent way that can easily reconcile with the overnight SONIA rate published by the BoE. For example, to calculate the compounded interest rate for a three month period requires approximately 60 data inputs (the overnight SONIA rates for each working day of the interest period) and there are a number of different ways in which the precise calculation can be performed that might create difficulties without universal acceptance of established conventions.
It is to overcome this (primarily operational) challenge, that the BoE plans to publish a “golden source” SONIA-linked index from which bespoke rates can more readily be calculated.
2. BoE consultation on publication of daily “screen rates” for specific period averages of compounded SONIA rates
The BoE recognises that some market participants are calling for it to go a step further and publish daily “screen rates” for one or more specific period averages – for example 6-month, 3-month or 1-month compounded SONIA rates – so that the agent need perform no calculation at all; it can simply use the relevant screen rate. It is therefore consulting on whether there is sufficient market consensus on which set of period averages it should publish as screen rates, or whether the number of different screen rates which would need to be published to meet the market’s needs would create the risk of confusion and undermine the very certainty which the BoE is seeking to achieve.
In our view, while the index is helpful, it is the BoE’s second initiative to consult on publishing a daily “screen rates” for specific period averages which is what the market really wants and needs. The need for a term replacement remains as strong in the market as ever, and the comments from the BoE underline the regulatory commitment to force the transition, irrespective of the state of market readiness, and so the development of a term replacement will continue to rest with the market.
Even if ultimately the BoE decides only to publish a SONIA-linked index, this represents a step forward in terms of helping the bilateral and syndicated loans market move to using RFRs. A number of the barriers to the Loan Market Association (LMA) producing a RFR facility agreement, or even a hard-wired fallback to RFR document, stem from the fact that the market has not yet agreed on the method of calculation of the compounded average RFR. This proposal from the BoE largely dispenses with these concerns, aids standardisation and provides a screen rate for agents to use, assuming the market is willing to accept a single standardised source calculated on this basis.
In terms of litigation risk, the simple point is that if publication of the SONIA-linked index (or daily screen rates for specific period averages) reduce the volume of new cash products written in LIBOR, then that will mean a reduction in the volume of risky LIBOR-linked products in existence on cessation of LIBOR.
However, a number of questions still remain. In particular, even if the operational challenges are removed by publication of a SONIA-linked index (or daily screen rate) in order for parties to transition legacy LIBOR-linked contracts to SONIA, they will have to agree the quantum of the spread adjustment which mitigates or eliminates the value transfer between them. As discussed in our article in the Journal of International Banking Law and Regulation on the types of litigation which may arise on LIBOR discontinuation, there is no procedure to amend legacy loan facilities on an industry-wide basis and there are clear obstacles for achieving effective adoption of revised fallbacks on a bilateral basis. As such there still remains a risk of significant volumes of legacy loan facilities which are not amended before LIBOR cessation takes place, notwithstanding this development.
3. BoE to increase haircuts on LIBOR-linked collateral it lends against from October 2020
And so to the stick. The BoE currently makes funding available to firms as part of its normal market operations against a wide set of eligible collateral, but applies a “haircut” to that collateral to protect against possible falls in its value in the period between a counterparty default and collateral sale.
The current average haircut is just under 25%, but from Q3 2020 the BoE will progressively increase the haircuts on LIBOR-linked collateral – increasing to 35% in October 2020, to 65% in June 2021 and, at the end of 2021, to 100% (i.e. effectively rendering such collateral ineligible). In addition, LIBOR-linked collateral issued after October 2020 will be ineligible for use.
The BoE says this graduated approach will give firms the time they need to replace that collateral with RFR alternatives, ensuring their borrowing capacity is maintained while also protecting public funds. The intended effect is clear, the higher the haircuts the less attractive the collateral, and so this step is likely to further wean firms from holding stocks of LIBOR linked assets and continue the evolution towards RFRs.
This development presents an interesting and novel litigation risk in relation to LIBOR discontinuation for market participants. If firms are unable to transition a sufficient amount of their book of LIBOR-linked products to SONIA in time, then it will affect their borrowing capacity at the BoE. Given that the BoE’s lending operations are designed to provide liquidity support to market participants experiencing either a predictable liquidity need or responding to an idiosyncratic or market wide liquidity shock, impairing the availability of effective liquidity at a possibly critical moment for firms could be significant.