LIBOR is being overtaken: Will it be a car crash?

It is clear that, from 2021, LIBOR (at least as we know it) will cease to exist. Given its prevalence as an interest rate benchmark in contracts across multiple markets and jurisdictions, its demise raises questions about the litigation risks which parties to such contracts may face. Herbert Smith Freehills LLP have published an article in the Journal of International Banking Law and Regulation on the types of litigation which may arise, and some of the legal arguments which might be deployed. The full article is available here: LIBOR is being overtaken: Will it be a car crash? (2019) 34 J.I.B.L.R.

In our article we explore the sorts of claims which may arise by reference to the four markets which are most affected by the transition from LIBOR: (A) the loan market; (B) the derivatives market; (C) the bond market; and (D) the securitisation market. We also look at the potential regulatory consequences (both to the financial institution and to the applicable senior manager) of failing to put in place appropriate transition plans, and to demonstrate delivery of that plan going forwards.

SONIA in the driving seat

Since the FCA confirmed that it will no longer compel banks to provide quotes for LIBOR after 2021, steps have been ongoing to determine an appropriate alternative risk free rate ("RFR"). In the UK, the Sterling Overnight Interbank Average Rate ("SONIA") has emerged as the preferred replacement rate and the Bank of England's Working Group on Sterling Risk-Free Reference Rates has an objective to ensure a transition from sterling LIBOR to SONIA by the end of 2021.

However, such a transition is not a matter of a straightforward switch and differences in the way the two rates operate will lead to various challenges. One of the fundamental differences (and a likely trigger for disputes) is that, since it is an overnight rather than a term rate, SONIA rates will be inherently lower than LIBOR, meaning that there will be potential value transfers if LIBOR is simply switched out for SONIA.

Will it be a car crash? The litigation risks

A. The loan market      

There is no procedure to amend legacy loan facilities on an industry-wide basis. Although the Loan Market Association ("LMA") produces a suite of recommend form facility agreements, they are not drafted to be adopted wholesale and are negotiated on a deal-by-deal basis. The process of amendment is also complicated by the fact that amendment clauses often require the unanimous consent of syndicate lenders and borrower. As such there are likely to be a significant number of unamended legacy loan facilities when LIBOR is finally discontinued.

For legacy contracts which are not amended, the existing LMA documentation provides a variety of options for parties to select from in a waterfall of fall-backs should LIBOR be unavailable:

1. Interpolated Screen Rate: LIBOR for a different tenor for the relevant interest period (which would not assist in a world where LIBOR has ceased for all currencies and maturities);
2. Historic Screen Rate: Last available LIBOR rate for the same currency and relevant interest period (effectively converting a floating rate to a fixed rate);
3. Reference bank rate: A rate calculated by the facility agent using rates provided by certain reference banks (which would rely on reference banks being willing/able to submit rates); or
4. Cost of funds calculation: Each lender's self-certified cost of funding its participation in the loan (a subjective calculation which is likely to be impractical to calculate and evidence).

Whatever the applicable fall-back, it is very plausible that in practice there will be a material 'winner' and a material 'loser' as a result of the transition from LIBOR. The 'losing' party may seek to avoid the fall-back position by deploying some or all of the arguments below.

• Contractual interpretation: That references to LIBOR should be read as references to an alternative reference rate, e.g. SONIA.
• Implied terms: That courts should imply adoption of an alternative reference rate.
• Frustration: That the contract is impossible to perform as a result of LIBOR discontinuation and should be brought to an end.

Arguments on the basis of contractual interpretation, implied terms and frustration are likely to be challenging in the context of LMA agreements given that they expressly provide for a range of (albeit not particularly commercial) fall-back mechanisms. Another battleground for litigation in this scenario will of course be challenges to the determination of the fall-back rate under the contractual provisions. Whilst the 'historic rate' will be straightforward to apply, the 'cost of funds' calculation is much more difficult in practice and vulnerable to challenge.

For amended legacy contracts and new contracts, the LMA has produced a 'Recommended Revised Form of Replacement Screen Rate Clause'^[1] which has been in use since the end of May 2018. The revised form of the clause provides for borrower and Majority Lender (i.e. two-thirds of the syndicate lenders) consent to be sufficient, rather than unanimous lender consent, to make amendments to the finance documents to facilitate the replacement of LIBOR with a replacement RFR, on the occurrence of a specified trigger event. However, there is plainly scope for dispute about the operation of the consent clause, application of the triggers and the operative provisions in the revised clause (which envisage, but do not actually provide, an adjustment mechanism to reduce/eliminate the transfer of economic value on transition from LIBOR).

B. The derivatives market

Unlike the multi-lateral syndicated loan market, derivatives are bilateral in nature. Moreover, the ubiquitous adoption by market participants of the ISDA documentation allows for an industry-wide mechanism to be used for the adoption of amendments. However, although ISDA intends to publish a backward-looking protocol (or set of protocols) which would allow parties to legacy contracts to elect to incorporate revised fall-backs, this will not provide a complete answer as it will only apply if both parties actively sign up to the protocol. Given the vast numbers of derivatives contacts, there remains a significant risk that a number are likely to be left unamended.

For unamended legacy contracts, the existing fall-back in the ISDA documentation provides for the calculation of an arithmetic mean of quotations (known as the 'dealer poll' method). This requires the Calculation Agent to obtain at least two quotations from certain specific Reference Banks or alternatively "major banks in London". The fall-back options are therefore much more limited in the ISDA documentation than they are in the LMA documentation and there is a real danger that, in circumstances where banks are not prepared to provide quotations, the courts will be left to determine the solution. Again, there are likely to be 'winners' and 'losers', and some of the potential arguments we have identified are set out below.

• Contractual interpretation/implied terms: Similar issues arise as under the LMA language.
• Contractual machinery: An alternative approach may be for the court to treat this scenario as an instance where the contractual machinery has broken down. In one line of authority, the courts have considered that, where that machinery was merely "subsidiary and inessential" to the parties' agreement on a "fair and reasonable" price^[2], the courts may substitute an alternative machinery to achieve the objective of setting an interest rate and avoiding the destruction of the parties' bargain. However, it may be difficult to characterise the selection of LIBOR as the reference rate as a "subsidiary and inessential" element of an ISDA agreement (particularly in the context of an interest rate swap with fixed vs. LIBOR payment legs) and it is difficult to see LIBOR as representing a measure which can be objectively determined by the court.
• Force majeure: There is no 'force majeure' clause under the 1992 ISDA Master Agreement^[3], but there is in the 2002 version. Under 2002 ISDA a Force Majeure Event will trigger the early termination provisions, end open trades and provide for the payment of a 'close out' amount. Whilst the discontinuation of LIBOR may not be a classical 'force majeure' event, it may provide a more attractive solution to the court than frustration. This is because it would provide a more orderly set of economic consequences, although it would not address the disruption caused by the lack of contractual continuity (and the need for replacement hedges on a vast scale).
• Frustration: This argument is more forceful under 1992 ISDA in the absence of a 'force majeure' clause. If no rates are submitted/obtained in the dealer poll method it could be argued that the contract has been frustrated and the parties are discharged from future obligations (without the need for any close out payment, in contrast to the situation for 'force majeure'). This argument has more scope in the derivatives market than under the LMA, because the ISDA legacy fall-backs are so limited.

For new contracts, ISDA's proposed approach is to publish a Supplement, so that any transactions entered into after the date of the Supplement will automatically incorporate the revised fall-back language (and an opt-in protocol for parties who wish to amend legacy contracts). ISDA's proposal is that the replacement RFR will be calculated by an independent third party and published on a screen so that it can be accessed in the same way that LIBOR currently can. This mechanism has the potential to limit the scope for another avenue of litigation to develop as a result of individual market participants calculating their own rate (but may expose the third party vendor to the risk of liability for any errors in its application of the chosen methodology). As with the LMA revised fall-back language, there is also the potential for disputes arising from the triggers to activate the fall-back and calculation of the spread adjustment (to reduce/eliminate the transfer of economic value on transition from LIBOR).

C. The bond market

The documentation for the bond market is bespoke to individual transactions. As such, there is no model language to analyse legacy bonds already in issue and there is no straightforward mechanism for amendment. However, based on a review of precedent transaction documents, common fall-back language relies on:

1. Reference bank rate: A variation of the 'reference banks' methodology which is found in both the LMA and ISDA clauses described above; and/or
2. Last available published LIBOR rate: There are particular risks that would arise from the effective conversion of a floating rate instrument to a fixed rate note. That would create inevitable disruption in the market for those bonds because many investors (for example, money market funds) would no longer be able to hold the notes if they became fixed rate instruments, forcing sales and likely driving prices downwards, possibly permanently.

Potential candidates for disputes include:

• Issuer liability-based clams: Along with mis-selling below, this is the most likely candidate for investor claims if the rate reverts permanently to the last available published LIBOR. Claims may allege that the features and risks of the notes (in particular the fall-back mechanisms and their operation in the event of LIBOR discontinuing) were not adequately disclosed to investors who subscribed for them.
• Mis-selling claims: Claims for breaches of advisory duties and/or concurrent tortious duties of care in recommending the purchase of the notes, against financial advisors or other intermediaries involved in the sale of the bonds to investors.
• Enforcement mechanism: Payments made on the basis of a replacement rate may engage the enforcement mechanisms in the documentation if that rate is successfully challenged.

For new bond issues, a commonly seen solution is that, upon a trigger event, the issuer will appoint an 'independent adviser' to determine the replacement rate and any necessary adjustment to reduce or eliminate any economic prejudice or benefit to noteholders. However, it remains to be seen if there are third parties who are willing to step in and accept such roles and it may be arguable that any financial adviser taking on that role will assume responsibility to investors to exercise those functions with reasonable care, and the potential for claims if that duty is not discharged.

D. The securitisation market

As with the bond market, securitisation documentation is bespoke and trustees or agents (on behalf of the issuer) are likely to face challenges in obtaining the necessary consent from the holders of the instruments (particularly at each different tranche of the securitisation structure) to make amendments to legacy documentation which would provide improved fall-backs to apply when LIBOR discontinues.

There is no uniform language in legacy instruments across the different products in the securitisation market. However, many public deals involve fall-backs based on:

1. Reference bank rate; or
2. Last available published LIBOR rate.

This will potentially give rise to similar disputes to those identified for the bond market. However:

• Contractual interpretation/implied terms: There may be different arguments to make here because conversion to a permanent fixed rate may give rise to particular commercial difficulties in circumstances where it results in mismatches between the interest rates paid to holders of the notes on the one hand, and the interest rates paid to the issuer on the underlying collateral within the structure.
• Claims against the trustee: It is not uncommon for the trust documents to provide the trustee with broadly defined powers to name a successor to LIBOR. The trustee's exercise of its power (or a failure to exercise it) may be a source of litigation to arise.

The Securitisation Division of the Association for Financial Markets in Europe has published wording for newly issued instruments, which includes a range of triggers. The proposed language introduces a mechanism which allows the issuer (through its agent) and/or the bond trustee to convert to an alternative benchmark rate without the need for requisite approvals from noteholders, subject to obtaining various rating agency confirmations and a negative consent mechanism. This raises obvious questions about the standard to which the agent or trustee will be held in exercising this discretion (or in not doing so). The precise operation of the negative consent mechanism is a further source of litigation, particularly if it could be argued that the senior noteholders' interests are being pursued illegitimately by the modification to the detriment of the more junior noteholders' interests. The role of credit rating agency confirmations also raises the possibility of litigation against the rating agencies, possibly deploying the causes of actions provided by CRA3^[4].

Regulatory consequences

The UK regulators have emphasised to the market that transitioning from LIBOR is an important issue, as illustrated by the Dear CEO letter issued to large banks and insurance companies on 19 September 2018. Inadequate LIBOR transition arrangements could give rise to:

• Enforcement action against a firm for a breach of:
  • FCA Principles for Businesses, most obviously Principle 2 (Skill, care and diligence) and Principle 3 (Management and Control);
  • PRA Fundamental Rules (FR2 and FR5); and
  • The related rules in the Senior Management Arrangements, Systems and Controls section of the FCA's Handbook.
• There are clearly also potential ramifications for individuals, particularly the individuals given the responsibility for implementation of the firm's transition plan. Awareness of the potential for individual accountability has been heightened since the introduction of the Senior Manager and Certification Regime for banks in 2016.

[1] Latest version dated 16 October 2018.

[2] Sudbrook Trading Estate Ltd. v Eggleton and Others [1983] 1 AC 444. At [461F].

[3] Unless the ISDA Illegality/Force Majeure Protocol has been subsequently followed.

[4] The Credit Rating Agency Regulations, Regulation (EU) No 462/2013. See further: CRA3 and the Liability of Rating Agencies: Inconsistent Messages from the Regulation on Credit Rating Agencies” (2013) 7(4) Europe, Law and Financial Markets Review