This judgment was handed down remotely at 2:00pm on 15 October 2024 by circulation to the parties or their representatives by e-mail and by release to the National Archives.
Sir Julian Flaux C and Mr Justice Foxton:
- This case has been brought under the Financial Markets Test Case Scheme established by CPR Practice Direction 63AA to determine the effect of the cessation of publication of the London Interbank Offered Rate ("LIBOR") on perpetual preference shares which provide for the payment of dividends by reference to that rate.
The Parties
- The Claimant ("SC") is an international bank listed on the London Stock Exchange and with its head office in London. In 2006, it issued USD 750 million preference shares ("the Preference Shares") to provide for what at the time was classified for regulatory purposes as "Tier 1" capital.
- The sole registered shareholder of the Preference Shares is the First Defendant ("GNL") as nominee for JPMorgan Chase Bank NA ("the Depository"). The Depository in turn issued American Depository Shares ("ADSs"). It was the holders of those ADSs from time-to-time who held the economic interest in the Preference Shares. GNL did not take an active role in these proceedings.
- The Second to Fifth Defendants ("the Funds") are some of the holders of the ADSs. They are members of what has been referred to as the "Ad Hoc Group" of ADS holders, which group the court has been told represents the holders (directly or beneficially) of just over 10% of the ADSs by value. The precise number of ADS holders who support the position taken by the Funds in these proceedings does not matter, because none of the legal and expert issues before the court are affected by it.
The Preference Shares
- As is well-known, banks and financial institutions such as SC are required to satisfy certain capital ratios set by regulators as part of their oversight of financial markets, and which in turn reflect international requirements set by the Basel Committee on Banking Supervision. Different ratios apply to different types (or tiers) of capital.
- On 8 December 2006, SC issued the Preference Shares for the purpose of raising Tier 1 Capital. The Preference Shares were issued in accordance with Article 7(A) of SC's Articles of Association, which provide that SC may issue a series of preference shares "with such rights … as the board may determine in the resolution approving the issue of such shares." The resolution in question is recorded in the minutes of the relevant committee of SC's board of 22 November 2006, at which the terms on which the Preference Shares were to be issued, and of the offering circular by which the ADS would be offered for sale ("the Offering Circular"), were also approved.
- There are two varieties of ADS, those offered to domestic US investors (so-called "Rule 144A" investors) and those offered to international investors (Regulation S investors).
- In the usual manner, the Offering Circular set out the terms of the ADSs, and sought to identify the key risks which investment in them entailed. For present purposes, it is sufficient to note the following:
i) The Preference Shares had a nominal value of USD 5 each, and were issued at a premium of USD 99,995 fully paid cash (such that the total paid-up amount of each Preference Share was USD 100,000).
ii) The net proceeds of the Preference Share issue were to be used "for the general business purposes of the Group, which may include acquisitions."
iii) The Preference Shares ranked pari passu inter se and pari passu with existing preference shares issued by SC.
iv) The USD 750 million preference share issue involved 7,500 ADSs.
v) The Preference Shares and ADSs were perpetual, which is to say that they had no maturity date.
vi) SC, but not the ADS holders, had the option to redeem the Preference Shares "subject to the Articles, provisions of applicable law and to the prior consent of the FSA (if such consent is required)" in whole or in part on 30 January 2017, and at 10-year intervals thereafter.
vii) The ADSs and associated contracts were governed by the laws of the state of New York but the Preference Shares by the laws of England and Wales. It was common ground at this hearing that the issues before the court were to be determined by reference to the law of England and Wales.
viii) Dividends were payable at a fixed rate of 6.409% per annum, semi-annually, until 30 January 2017, and thereafter at a floating rate of "1.51% plus Three Month LIBOR". We will consider the definition of that term in more detail below.
ix) However, in contrast to a coupon on a note or bond, dividends were payable at the discretion of the board, with the board not being permitted to pay dividends if in its opinion "such payment would exceed available distributable profits or breach capital adequacy requirements." If the dividend was not paid in full on the relevant date, then SC could not pay a dividend on its ordinary shares for a period of one year thereafter (together with certain other restrictions for varying periods). However, the dividends on the Preference Shares were non-cumulative, such that if no dividend was payable on one Dividend Payment Date, the amount of the dividend payable on that date could not be accumulated with that payable on the next Dividend Payment Date.
- Reverting to the definition of "Three Month LIBOR", this provided:
"'Three Month LIBOR' means the three month London interbank offered rate for deposits in US dollars which appears on page 3750 of Moneyline Telerate as of 11:00 a.m., London time, on the second business day in London prior to the first day of the relevant Dividend Period…;
provided that, if at such time, no such rate appears or the relevant Moneyline Telerate page is unavailable, it shall mean the rate calculated by the Company as the arithmetic mean of at least two offered quotations obtained by the Company after requesting the principal London offices of each of four major reference banks in the London interbank market, to provide the Company with its offered quotation for deposits for three months in US dollars commencing on the first day of the relevant Dividend Period to prime banks in the London interbank market at approximately 11:00 a.m., London time, on the second business day in London prior to the first day of the relevant Dividend Period and in a principal amount that is representative for a single transaction in US dollars in that market at that time;
provided further that if fewer than two such offered quotations are provided as requested, it shall mean the rate calculated by the Company as the arithmetic mean of the rates quoted at approximately 11:00 a.m., New York time, on the second business day in New York prior to the first day of the relevant Dividend Period, by three major banks in New York selected by the Company for loans for three months in US dollars to leading European banks and in a principal amount that is representative for a single transaction in US dollars in that market at that time;
provided however that if the banks selected by the Company, are not quoting as mentioned above, it shall mean three month US dollar LIBOR in effect on the second business day in London prior to the first day of the relevant Dividend Period."
- It will be apparent that the definition contains a primary means of ascertaining LIBOR, with three alternatives if and to the extent that the prior means cannot be operated. The parties referred to those alternatives as the First Fallback, the Second Fallback and the Third Fallback respectively.
- We understand the following matters to be common ground so far as any redemption of the Preference Shares is concerned:
i) The acquisition by a company of its own shares is subject to a number of restrictions in the Companies Act 2006 ("CA 2006").
ii) Any right to redeem preference shares must be conferred on the issue of the shares rather than by way of subsequent amendment: ss.684(1) and 685(3) CA 2006. That is so even if a subsequent amendment has the unanimous consent of the shareholders (Re St James Court Estate [1944] Ch 6).
iii) The terms, conditions and manner of redemption must be set out before the shares are allotted: ss.685(1) and (3) CA 2006.
iv) A public company (such as SC) must finance the redemption from distributable reserves or the proceeds of a fresh share issue: ss.687(1) and (2) CA 2006.
- The regulatory classification of the Preference Shares was changed from Tier 1 to Tier 2 capital.
LIBOR
- Before exploring the definition further, it is helpful briefly to summarise the history of LIBOR.
- The evidence before the court indicates that LIBOR was first conceived in 1969 for use in connection with a proposed loan by a syndicate of banks to the Shah of Iran. The loan adopted a floating interest rate, recalculated on a periodic basis by reference to the weighted average borrowing costs of each reporting bank in the syndicate, with a fixed spread for profit.
- LIBOR soon came to be adopted in the derivatives market, and in contracts by which banks raised funds for their own purposes.
- In 1986, the British Bankers' Association ("the BBA") took over the administration of LIBOR and began publishing LIBOR rates for different currencies and different maturity terms (or "tenors").
- Between 1983 and 2007, the volume of debt issuances using USD LIBOR-indexed rates increased from USD 255 million to USD 1,057 billion. In 2006, when the Preference Shares were issued, floating and variable rate debt issuances using USD LIBOR were USD 1,038 billion, or 78% of all such issuances.
- In the period up to 2016, LIBOR was based on the cost of interbank unsecured lending – i.e. what it would cost one bank to borrow from another bank on an unsecured basis. A qualitative survey was taken from a group of between 11 and 18 panel banks. Up to 1998, panel banks were asked "at what rate do you think interbank term deposits will be offered by one prime bank to another prime bank for a reasonable market size today at 11am." In 1998, the question became:
"At what rate could you borrow funds were you to do so by asking for and then accepting interbank offers in a reasonable market size just prior to 11am?"
- The panel banks were chosen by the BBA based on the BBA's perception of the scale of their activity, their expertise in the currency concerned and their credit standing.
- The panel banks' responses were "trimmed" by removing responses at the top and bottom of the range – if there had been 15 responses, the four highest and four lowest submissions were removed – and an arithmetic mean derived from the remaining seven returns. As a qualitative assessment of what it would cost the responding bank to borrow money, LIBOR necessarily reflected the credit risk in such a transaction.
- The financial crisis that began in April 2007 revealed a number of flaws in the LIBOR rate. Allegations were made of collusion between panel banks, with rates being arrived at not by way of a genuine estimate of the responding bank's borrowing cost, but in an attempt to present the bank's creditworthiness in a more favourable light, or to influence the published rate for the purpose of profiting from LIBOR-linked transactions in which the submitting bank was involved. The US Department of Justice began a criminal investigation into the manipulation of LIBOR, which culminated in a series of LIBOR panel banks being fined for inappropriate conduct in relation to their LIBOR returns.
- Concerns in relation to the LIBOR rate led to regulatory and industry initiatives with a view to identifying floating rates which could be used as alternatives to LIBOR. Those efforts were prolonged and comprehensive, and produced a vast quantity of material, some of which was placed before us. Both SC and the Funds relied upon this material. We have given a longer summary of the relevant events in the Appendix to this judgment, with references in the main body of the judgment to paragraphs in that Appendix taking the form [A1], etc.
- By way of a brief summary:
i) The Federal Stability Board, an international body that monitors and makes recommendations about the global financial system, recommended the development of nearly risk-free reference rates, embedding no or only small amounts of credit risk, given the wholesale decline in unsecured short-term funding by banks.
ii) The Federal Reserve System of the United States ("the Fed") and the Federal Reserve Bank of New York convened the Alternative Reference Rates Committee ("ARRC") to "identify a set of alternative reference rates that are more firmly based on transactions from a robust underlying market and that comply with emerging standards". The ARRC initially consisted of representatives from 15 large global interest rate derivative dealers, the Fed, the SEC and the US Treasury.
iii) The ARRC undertook an extensive programme of market consultation and produced a number of reports and analyses which considered and assessed possible alternative reference rates.
iv) By June 2017, the ARRC had recommended a new risk-free rate as the appropriate alternative reference rate, being a daily rate for overnight lending secured by US Treasuries: the Secured Overnight Funds Rate or SOFR.
v) The Fed released a range of historical indicative data for the period from August 2014 to March 2018 based on modelled, pre-production estimates of SOFR relying on the same underlying data and methodology as the published rate.
vi) In addition, historical survey data was released by the Federal Reserve Bank of New York to assist in the calculation by the Fed of a historical proxy for overnight SOFR going back to 1998.
vii) The International Swaps and Derivatives Association ("ISDA") initiated a market-wide consultation to consider how to address the difference (or "spread") between SOFR and LIBOR. The overwhelming majority of respondents expressed the view that a spread adjustment based on a historical median over a five-year lookback period was appropriate.
viii) After conducting its own consultation process, the ARRC also endorsed the use of a fixed rather than dynamic spread adjustment and endorsed the spread adjustment proposed by ISDA ("the ISDA Spread Adjustment") concluding that: "if economic conditions remain relatively calm, the spread adjustments to be recommended by the ARRC could have very little associated error". It was also noted that "average annualized errors are also likely to be smaller the longer the remaining maturity of the loan or security being referenced at the time that LIBOR stops, because any larger errors are more likely to 'average out' over the remaining life of the instrument."
ix) To provide forward term SOFR rates ("Term SOFR") the ARRC recommended use of the Term SOFR rates published by the Chicago Mercantile Exchange Group Benchmark Administration ("CME Term SOFR"). CME Term SOFR is a forward-looking rate calculated on a futures basis by reference to trading in derivatives on the CME, and reflects market expectations for SOFR in the future.
x) On 15 March 2022, the US Congress passed the Adjustable Interest Rate (LIBOR) Act which contains a series of default rules for legacy contracts which did not provide for the use of a clearly defined and practical replacement benchmark for LIBOR for the period after 30 June 2023 in contracts which were subject to US law. The Act made provision for the Board of Governors of the Federal Reserve System to fix replacement rates for those contracts.
xi) On 26 January 2023, the Board exercised the power granted by the Act to stipulate a replacement rate for non-derivative and non-consumer transactions on USD three month LIBOR of three month CME Term SOFR plus the ISDA Spread Adjustment.
- So far as the position in the United Kingdom is concerned:
i) The Wheatley Review of September 2012 recommended a series of reforms to the process for producing LIBOR.
ii) By way of implementing some of those recommendations, administration of LIBOR was handed over from the BBA to the ICE Benchmark Administration Limited ("IBA"), who began publishing "ICE LIBOR" in 2014 and who were regulated by the FCA.
iii) In 2016, the IBA identified issues with the publication of LIBOR arising from the substantial decrease in the interbank unsecured loan market in the aftermath of the financial crisis.
iv) That reduction or "thinning out" of the interbank unsecured lending market led the IBA to update the definition of LIBOR in 2016 to include unsecured lending between banks and non-financial corporations.
v) On 27 July 2017, Andrew Bailey, the Chief Executive of the Financial Conduct Authority ("FCA"), gave a speech stating that the absence of an active underlying market raised a serious question about the sustainability of LIBOR benchmarks based on those markets and warning market participants not to rely indefinitely on reference rates which did not have active underlying markets to support them.
vi) Over a sustained period, UK regulators encouraged market participants to transition from LIBOR, warning that LIBOR might cease to be published at short notice.
vii) On 5 March 2021, the FCA announced that the IBA would stop publishing 24 (non-USD) currency and tenor LIBOR settings after December 2021, with synthetic sterling and yen rates continuing for a period (i.e. a rate calculated using market data rather than based on a survey of panel banks).
viii) In June 2021, the Bank of England and the FCA announced their support for the US initiative to move from USD LIBOR to a SOFR rate.
ix) In November 2022, the FCA carried out a further consultation on whether it should require publication of 1-, 3- and 6-month USD LIBOR settings on a synthetic basis until the end of September 2024, "as well as on the appropriate methodology of constructing such synthetic US dollar LIBOR settings and what use of them should be permitted." The FCA reported that it had decided it "should follow the same approach for US dollar LIBOR settings as for sterling and yen LIBOR settings", using the CME Term SOFR rate which it described as "robust and transparent." The report also stated:
"As we have said previously, while in our view synthetic LIBOR settings are a fair and reasonable approximation of what LIBOR might have been had it continued to exist, they are not representative of the markets that the original LIBOR settings were intended to measure."
x) The USD LIBOR bank panel ceased to exist on 30 June 2023.
xi) The FCA exercised regulatory powers to require the IBA to publish synthetic rates for 1, 3 and 6-month USD LIBOR to support a transition from LIBOR effective from 00.01 on 1 July 2023. The FCA provided that those rates should be "based on the relevant CME Term Reference Rate and the corresponding ISDA spread adjustment". The FCA stated that it was satisfied that this was a "fair and reasonable approximation of the value panel-bank LIBOR would have had".
xii) At the end of September 2024, in the course of the hearing of this case, synthetic USD LIBOR ceased to be published.
SC's response to the abolition of LIBOR
- There has been no challenge to SC's evidence that on the cessation of the publication of synthetic USD LIBOR at the end of September 2024, it is no longer possible to operate the First and Second Fallbacks in the definition of Three Month LIBOR because banks have not and will not provide the required quotations. On the Funds' construction, the Third Fallback cannot be operated either.
- On 8 November 2022, SC launched a "consent solicitation process" to amend the dividend rate for the Preference Shares by special resolution. SC sought a consensual amendment which would have replaced 3-month USD LIBOR with three month Compound SOFR (i.e. a rate which takes historical daily SOFR values and compounds them over the required term) plus the ISDA Spread Adjustment (a rate very similar, but not identical, to the rate now proposed).
- At a meeting on 4 January 2023, the proposed change failed to achieve the requisite 75% majority. Only 67% of the votes cast, which in turn represented 14% of the ADSs, were in favour of the proposed amendment.
- On 19 June 2023, SC announced that dividends would be calculated by reference to 3-month Synthetic LIBOR (i.e. CME Term SOFR plus the ISDA Spread Adjustment) for the dividend period on 31 July 2023.
- On 12 April 2024, SC issued a Claim Form in the Financial List with GNL as the sole defendant, seeking declarations concerning the rate by reference to which the dividends payable on the Preference Shares should be calculated for the dividend periods commencing on or after 30 October 2024. That application was supported by an expert report from Dr Faten Sabry which expressed the view that CME Term SOFR plus the ISDA Spread Adjustment was the appropriate rate to use.
- The Funds indicated in correspondence that they wished to participate in the proceedings, and to oppose the relief sought by SC. On 10 May 2024, Mr Justice Foxton approved an order by consent joining the Funds to the proceedings as the Second to Fifth Defendants. Mr Justice Foxton held an urgent directions hearing on 15 May 2024 at which he gave the Funds permission to serve responsive evidence (which in due course was served in the form of a report from Dr Albert Metz).
- On 14 June 2024, Mr Justice Foxton made an order moving the case into the Financial Markets Test Case scheme, to be heard by a two-judge court including a judge of the Court of Appeal. The case was in due course listed before this constitution which heard the case as, in effect, a Divisional Court.
The Parties' Cases
SC's case
- SC's primary case is that the phrase "three month US dollar LIBOR in effect" in the Third Fallback should be construed as "a rate that effectively replicates or replaces three month USD LIBOR". Pausing there, we would note the distinction between:
i) "replication", which would test the suitability of the rate to be used in the place of three month USD LIBOR by how far it produced the same outcome as three month USD LIBOR; and
ii) "replacement", which would test the suitability of the rate to be used by reference to whether that rate had come to supplant three month USD LIBOR in the market.
- SC's alternative case is that a term should be implied into the terms governing the Preference Shares, as fixed by SC's board under the power provided for in the Articles of Association, "that allows SC to use a reasonable alternative rate once synthetic USD LIBOR settings cease to be published". The final implied term proposed by SC was as follows:
"A term should be implied in the definition of Three Month LIBOR in the Offering Circular that, where the express definition fails, SC should use a reasonable alternative rate to three month USD LIBOR".
- In either case, SC submits that a rate produced by taking CME Term SOFR and adding the ISDA Spread Adjustment meets the requirements for the alternative rate. We will refer to this as the Proposed Rate.
The Funds' case
- The Funds initially sought a declaration "that a term should be implied into the terms of the Preference Shares that if three month USD LIBOR ceases to be available the Claimant shall redeem the Preference Shares".
- The final form of the implied terms proposed by the Funds were rather more complex, in an attempt to cater for various criticisms of its case made by SC. The Funds proposed a two-stage approach.
- The "stage one" implied term was as follows:
"If three month USD LIBOR ceases to be available then, subject to the Companies Act and all other laws and regulations applying to the Company, to the Articles and to the prior consent of the FSA (if such consent is required, in which case, the FSA may impose conditions on the redemption), the Company shall, upon not less than 30 nor more than 60 days' notice, redeem the Preference Shares in whole on the second business day in London prior to the first day of the first Dividend Period that follows both the date on which three month US dollar LIBOR ceases to be available and the commencement of the Floating Rate Period (the "Impossibility Redemption Date"). There shall be paid on each Preference Share so redeemed the aggregate of: (i) an amount paid up on such share; and (ii) the dividend accrued for the period from, and including, the Dividend Payment Date last preceding the Impossibility Redemption Date to the date to, but excluding, the Impossibility Redemption Date."
- To the extent that the Preference Shares could not be redeemed in accordance with that implied term, the Funds advanced a "second stage" implied term as follows:
"If redemption in accordance with the above clause is, as at the Impossibility Redemption Date, unlawful under the Companies Act, any other applicable law and/or any applicable regulations, or if the FSA does not provide its prior consent,
[Alternative options:
1. The Company shall, until redemption becomes lawful, and/or is not in breach of regulations, and/or is permitted by the FSA, at which point the Preference Shares shall be redeemed, pay to the Holders of the Preference Shares, a sum as if it were a Dividend payable under the terms of the Preference Shares, with the dividend rate being equal to the last published LIBOR rate prior to the Impossibility Redemption Date, plus 1.51%.
2. The Company shall, until redemption becomes lawful, and/or is not in breach of regulations, and/or is permitted by the FSA, at which point the Preference Shares shall be redeemed, pay to the Holders of the Preference Shares, a sum as if it were a Dividend payable under the terms of the Preference Shares, with the dividend rate being equal to 6.409%.
3. Such other term as the Court sees fit.]"
The Applicable Principles for the Construction of Contracts
- The manner in which the contract came into being is unusual in this case, with a delegated power in the Articles of Association being exercised on a basis whereby there is single shareholder for the Preference Shares, with the economic interest in those shares being that of the beneficial owners of the ADSs from time-to-time. Nonetheless, there was no dispute that the issues before the court were essentially contractual, the Articles of Association of a company constituting a statutory contract between the company and the shareholders inter se (Cosmetic Warriors Ltd v Gerrie [2017] EWCA Civ 324, [19]), and the terms on which the Preference Shares are issued being, in effect, incorporated terms of that statutory contract.
- As would be expected, there was little dispute between the parties as to the general principles relating to the interpretation of contracts and the implication of contractual terms.
The interpretation of contracts
- So far as authorities on the interpretation of contracts are concerned, we were referred to a number of the "usual suspects", including the recent summary of the applicable principles given by the Supreme Court in Sara & Hossein Asset Holdings Ltd v Blacks Outdoor Retail Ltd [2023] UKSC 2, [2023] 1 WLR 575 at [29]:
"(1) The contract must be interpreted objectively by asking what a reasonable person, with all the background knowledge which would reasonably have been available to the parties when they entered into the contract, would have understood the language of the contract to mean.
(2) The court must consider the contract as a whole and, depending on the nature, formality and quality of its drafting, give more or less weight to elements of the wider context in reaching its view as to its objective meaning.
(3) Interpretation is a unitary exercise which involves an iterative process by which each suggested interpretation is checked against the provisions of the contract and its implications and consequences are investigated."
The implication of terms
- Both parties accepted that the relevant test is that set out in Marks & Spencer Plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd [2015] UKSC 72, [2016] AC 742:
i) An implied term must either be necessary to give business efficacy to the contract, meaning that the contract would lack commercial or practical coherence without the term ([17] and [21]) or be so obvious that it goes without saying ([16]).
ii) The term to be applied must be capable of clear expression ([18]), not contradict any express terms of the contract ([28]); and be reasonable and equitable, although a term which meets the previous requirements will almost certainly be reasonable and equitable ([21]).
Relevant "second order" principles
- These "first order" principles can be supplemented by "second order" principles addressing the application of the first order principles in particular contexts. There are three overlapping principles of this kind which are relevant here.
- First, this is a long-term contract – indeed, unless SC exercises its decennial option to redeem the Preference Shares, it would ordinarily be a perpetual one. While it has been said that "there are no special rules of interpretation applicable to such contracts" (Lord Steyn in Total Gas Marketing Ltd v Arco British Ltd [1998] 2 Lloyd's Rep 208, 218), that quotation continues:
"That is not to say that in an appropriate case a court may not take into account that, by reason of the changing conditions affecting such a contract, a flexible approach may best match the reasonable expectations of the parties."
- It has also been noted that a failure to address a specific issue in a long-term contract may be less significant than in a short-term contract. Butcher J noted of a long-term gas contract in Teesside Gas Transportation Limited v CATS North Sea Limited, Antin CATS Limited, Conocophillips Petroleum Company U.K. Limited, ENI UK Limited [2019] EWHC 1220 (Comm), [37]:
"It is … unrealistic to consider that the drafters will have envisaged every possible factual scenario which might arise under the contract over the course of the following three decades."
- In Mamidoil-Jetoil Greek Petroleum Co SA v Okta Crude Oil Refinery AD (No 1) [2001] EWCA Civ 406, Rix LJ noted the particular difficulties which arise in fixing a financial performance obligation (in that case a price) in a long-term contract. At [66]-[67], he referred to the decision of the House of Lords in Sudbrook Trading v Eggleton [1983] AC 444, (discussed at [48]-[49] below) and stated:
"The interest of Sudbrook Trading v. Eggleton for present purposes is that it may be said to demonstrate the difficulty for the parties of providing for a price when the event which requires the determination of the price will not happen until some time in the future: it is a price which, as under any long-term contract, cannot be determined at the time of contract. There the device was used of referring the valuation to a panel of valuers. But, as Lord Fraser said (at 483F/G) —
'In the ordinary case parties do not make any substantial distinction between an agreement to sell at a fair value, without specifying the mode of ascertaining the value, and an agreement to sell at a value to be ascertained by valuers appointed in the way provided in these leases. The true distinction is between those cases where the mode of ascertaining the price is an essential term of the contract, and those cases where the mode of ascertainment, though indicated in the contract, is subsidiary and non-essential.'
There is, in my view, implicit support here for the doctrine that in a commercial contract, which, when dealing with the future and sometimes the long-term future of necessity leaves certain matters such as price to be worked out over time, an arbitration clause assists the court to find sufficient certainty by means of the implication of what is reasonable. Which is not to say, that the court will not itself provide the dispute resolution machinery, even in the absence of an arbitration clause."
- Rix LJ's summary of the applicable principles at [69] included the following:
"iv) … [P]articularly in commercial dealings between parties who are familiar with the trade in question, and particularly where the parties have acted in the belief that they had a binding contract, the courts are willing to imply terms, where that is possible, to enable the contract to be carried out.
…
vi) Particularly in the case of contracts for future performance over a period, where the parties may desire or need to leave matters to be adjusted in the working out of their contract, the courts will assist the parties to do so, so as to preserve rather than destroy bargains, on the basis that what can be made certain is itself certain. Certum est quod certum reddi potest.
vii) This is particularly the case where one party has either already had the advantage of some performance which reflects the parties' agreement on a long-term relationship, or has had to make an investment premised on that agreement.
viii) For these purposes, an express stipulation for a reasonable or fair measure or price will be a sufficient criterion for the courts to act on. But even in the absence of express language, the courts are prepared to imply an obligation in terms of what is reasonable."
- The second of the three overlapping principles is most clearly encapsulated in Sudbrook Trading Ltd v Eggleton [1983] 1 AC 444, in which a lease granted an option to acquire a reversion in fee simple in the demised premises at a price which, in default of agreement, was to be determined by two valuers appointed by the parties. The valuation mechanism failed. In holding that the right to acquire the reversion did not fail with it, the House of Lords drew a distinction between provisions of a contract which are intended to define the substantive entitlement of the parties, and provisions which are in the nature of "machinery" intended to quantify that substantive entitlement. Where the machinery is a non-essential part of the contract, and is incapable of operation, the House of Lords held that the court can step in and itself perform the necessary exercise of quantification itself.
- In Didymi Corporation v Atlantic Lines and Navigation Co Inc (The Didymi) [1988] 2 Lloyd's Rep 108, 115, Bingham LJ reviewed various cases addressing the distinction drawn in Sudbrook Estates and identified the question for the court as follows:
"Does the provision in issue in this case relate to an essential term of the agreement or to the existence of any contract at all, or does it relate to a subsidiary and non-essential question of how a contractual liability to make payment according to a specified objective standard is to be quantified?"
- The last of those three overlapping principles emerges from authorities which specifically address how the task of interpretation is to be approached when the parties are confronted during the life of their contract with an event with contractual implications which they did not foresee:
i) In Debenham Retail Plc v Sun Alliance and London Assurance Co Ltd [2005] EWCA Civ 868, Mance LJ discussed the approach to be taken to the construction of a provision in a long-term lease in which the amount of rent payable was to be determined by reference to "the gross amount of total sales", in circumstances in which VAT had subsequently been introduced on the tenant's trading. At [27]-[28], he stated:
"To speak even of objective intention in such circumstances involves some artificiality. Even if we were judicial archaeologists, we would find in the wording of the lease negotiated in 1965 no actual or buried intention regarding VAT, since it was introduced in April 1973, and the regime in force in 1965 was the different purchase tax regime. But no-one suggests that the lease cannot or should not apply in the changed circumstances. We have to promote the purposes and values which are expressed or implicit in its wording, and to reach an interpretation which applies the lease wording to the changed circumstances in the manner most consistent with them.
In that exercise, I find compelling the submission that, although the method of collection is different—purchase tax being collected at the single (wholesale) stage in the production and distribution chain, and VAT being collected in instalments—both represent in effect "a tax on final consumer expenditure in the domestic economy"."
ii) In York CC v Trinity One (Leeds) Ltd [2018] EWCA Civ 1883, when a contract defined the amount payable by a developer by reference to a form of grant which ceased to exist after the contract was entered into, the Court of Appeal upheld the contract and once again distinguished between the payment obligation and the machinery for its quantification (at [34]), continuing at [38]:
"The clause provides that the developer should pay enough money so that the Council can provide equivalent affordable housing: the best the court can do is work out a roughly equivalent figure for that sum."
iii) In W Nagel (a firm) v Pluczenik Diamond Co NV [2018] EWCA Civ 2640; [2018] 2 CLC 938, Leggatt LJ stated at [31]:
"It is … commonplace for circumstances to arise which the parties to a contract did not foresee when the contract was made. When this happens, it does not follow that the contract ceases to be binding or ceases to apply. On the contrary, unless the change of circumstances is so radical or fundamental as to frustrate the contract by making it impossible to perform, the parties are held to their bargain. What the contract requires in the changed circumstances depends on its proper construction. As Chadwick LJ explained in Bromarin AB v IMD Investments Ltd [1999] STC 301, 310:
'It is not, to my mind, an appropriate approach to construction to hold that, where the parties contemplated event 'A', and they did not contemplate event 'B', their agreement must be taken as applying only in event 'A' and cannot apply in event 'B'. The task of the court is to decide, in the light of the agreement that the parties made, what they must have been taken to have intended in relation to the event, event 'B', which they did not contemplate. That is, of course, an artificial exercise, because it requires there to be attributed to the parties an intention which they did not have (as a matter of fact) because they did not appreciate the problem which needed to be addressed. But it is an exercise which the courts have been willing to undertake for as long as commercial contracts have come before them for construction. It is an exercise which requires the court to look at the whole agreement which the parties made, the words which they used and the circumstances in which they used them; and to ask what should reasonable parties be taken to have intended by the use of those words in that agreement, made in those circumstances, in relation to this event which they did not in fact foresee.'"
- These cases support an approach which, when a contract is required to be performed in (non-frustrating) circumstances which the parties did not foresee and for which they did not provide, seeks to ascertain the purpose or structure of the relevant aspects of the parties' bargain, and to adopt an interpretation which best serves or is most consistent with that purpose in the changed circumstances: in effect, a form of contractual cy-près.
- Such an approach, as well as cohering with the intentions of reasonable parties to long-term contracts, also gives effect to an important policy of English contract law which is reluctant to contemplate the failure of partly executed contracts merely because they do not address a particular circumstance or eventuality which has come to pass (Sir Kim Lewison, The Interpretation of Contracts 7th [8.135]; Mamidoil-Jetoil, [69](iv)]).
SC's interpretation argument
- Against that background, we turn first to SC's interpretation case.
- The expression "in effect" can be used in two senses: to mean "in fact" or "in practice" (cf. [51]), where it is a synonym for "effectively", or alternatively to mean "in force" or "in operation". SC asks the court to interpret the use of the phrase in the Third Fallback – "if the banks selected by the Company, are not quoting as mentioned above, it shall mean three month US dollar LIBOR in effect on the second business day in London prior to the first day of the relevant Dividend Period" – in the first sense. On that basis, it argues that, when LIBOR ceases to be published, the Third Fallback permits the use of a rate which is effectively LIBOR – either because it is the rate which is being used in transactions where LIBOR was previously used, or because it is closest in its outcome to 3-month USD LIBOR, or some combination of the two.
- However, we are satisfied that the expression "in effect" is used in the Third Fallback in its second sense. This contemplates a situation when no LIBOR rate is published on the relevant date, but a LIBOR rate first published on a prior date is treated as the effective LIBOR rate by the market (and so is "in effect" or operative) on the later date:
i) First, this is the natural meaning of the term "in effect" when used in the specific context of an identified point in time ("in effect on the second business day in London prior to the first day of the relevant Dividend Period").
ii) Second, this construction more naturally fits with the architecture of the definition of "Three Month LIBOR". The Third Fallback only comes into play if it is not possible to establish a rate at which a bank can borrow funds on the relevant date by obtaining quotations from a certain number of banks (either in London or New York); provisions which are themselves intended to provide a proxy to a published rate at the relevant time. The Third Fallback is the rate of last resort in the definition because it is not a "current" rate. On SC's construction, however, it is very difficult to understand why the Third Fallback should become available only after the First and Second Fallbacks have failed. On SC's approach, the First and Second Fallbacks are simply more restricted means of undertaking the same task which SC's construction of the Third Fallback contemplates.
iii) Third, the phrase "in effect" is used in the Offering Circular on a number of occasions in this second, temporal, sense: for example, the provision that the Preference Shares will "be subject to applicable UK tax laws and regulations in effect at the time of the exchange"; the definition of "Applicable Booking Regulations" as the "capital adequacy regulations, guidelines and policies then in effect of the FSA", and the definition of Articles as those "in effect from time to time".
- In short, the fallback provisions here are of the kind commonly found in financial instruments which used LIBOR as a reference rate, which, as noted in the Wheatley Review in 2012 at [5.30] "are intended to be used in the event of occasional operational problems, or other market disruptive events, which lead to LIBOR not being published in the usual manner."
- However, a number of features of the definition of Three Month LIBOR and the purpose of that provision in the context of the Preference Shares are highly relevant when considering SC's alternative "implied term" argument.
- First, the reference to LIBOR itself was a reference to an essentially judgmental and potentially imprecise mechanism for measuring the costs of unsecured bank borrowing over time, which measure was subject to significant change in the manner of its production over time. The published LIBOR rate was a means to an end, not Holy Writ in itself:
i) The LIBOR rate was arrived at from inputs from a varying number and identity of panel banks, who were asked a hypothetical and broad-textured question: initially "at what rate do you think interbank term deposits will be offered by one prime bank to another prime bank for a reasonable market size today at 11am" and from 1998 "At what rate could you borrow funds were you to do so by asking for and then accepting interbank offers in a reasonable market size just prior to 11am?"
ii) As the IBA noted in "ICE LIBOR Evolution" published on 25 April 2018, the effect of that change was that "each submission became a subjective determination of the rate at which a given Panel Bank could transact." It was the subjectivity of the responses to that hypothetical question which created the scope for the manipulation of submissions.
iii) The identity of the panel banks depended on the judgments of the BBA by reference to a range of criteria.
iv) The number of panel banks submitting to the BBA varied, and the fact that there might be a range of possible responses was recognised in the trimming exercise undertaken.
v) The IBA took over the administration of LIBOR in 2014.
vi) In 2016, LIBOR ceased to be an exclusively interbank rate and was extended to include lending between banks and non-financial corporations.
- LIBOR's function as a measure rather than a value in its own right is recognised in numerous statements about the rate: see for example [A21], [A27], [A31] and [A35]. Further, those statements (and the history of the fall of LIBOR) make it clear that LIBOR had become progressively less effective at measuring what it was intended to measure.
- Second, the three "Fallbacks" manifest the parties' common intention that issues relating to the publication of LIBOR should not prevent the continued operation of their contractual arrangements, even if this involved using alternatives which might involve an appreciably different outcome from their primary measure. Thus, the First and Second Fallbacks did not involve:
i) a rate published under the auspices of the BBA (or IBA/ICE);
ii) (necessarily) submissions from panel banks or those qualified to be such (or anything like the number of panel banks) or (in the case of the Second Fallback) London banks; and
iii) any exercise of "trimming" to remove outliers (such that it might have involved quotations from banks who, had they been panel banks, would have been "trimmed" as outliers).
The First and Second Fallbacks were also concerned with quotations for unsecured lending "in a principal amount that is representative for a single transaction in US dollars in that market at that time" rather than "a reasonable market size."
- Third, the various limbs of the definition contemplated a floating rate, with the preference being for a "real time" rate (rather than one reflecting the position at some earlier point in time, which only appears in the Third Fallback). This reflected the understandable desire to seek to "future-proof" the amount of the dividend by indexing to a rate which would provide a measure of movements in the cost of borrowing over time (a floating rate being chosen in the alternative to other "future-proofing" mechanisms, such as a contractual discretion on the part of one party to change the rate controlled by Braganza principles).
SC's implied term
- At the outset we note that neither party suggests that this is a case in which the absence of an express term addressing the cessation of publication of LIBOR means that "nothing is to happen" in that eventuality (cf. Attorney General of Belize v Belize Telecom [2009] UKPC 10, [17]). No one suggests that the Preference Shares would have commercial or practical coherence without the implication of a term to address the cessation of LIBOR. In these circumstances, each party contends that a term must be implied. The issue is what term.
- The bargain embodied in the Preference Shares was, on any view, a very long-term contract. In these circumstances, a flexible approach to its construction (embracing within that term both the interpretation of its express terms and the ascertainment of any implied terms) best matches the reasonable expectations of the parties (cf. [44]).
- We are also satisfied that the role of LIBOR in the Preference Shares is essentially to provide a measure which will link the amount of the dividend to the changing costs of borrowing over time, with the result that the provision is properly to be characterised as non-essential "machinery" for the purpose of determining what happens when LIBOR ceases to be published. That is clear both from the nature of LIBOR itself (see [58]) and from its treatment in the relevant contractual terms (see [60]). In these circumstances, when the mechanism cannot be operated the courts will generally be willing to imply an obligation by reference to what is reasonable to enable the contract to be carried out ([48]-[49]).
- Finally, as we have noted, it is clear from the terms on which the Preference Shares were issued that the parties did not intend issues with the availability of LIBOR to prevent the continued performance of the contract ([60]). Indeed, for reasons we explain at [83] below when addressing the Funds' implied term arguments, the fact that the Preference Shares were issued by way of Tier 1 capital for a regulated financial institution made it particularly important for the dividend mechanism to continue to operate even if LIBOR ceased to be published.
- In these circumstances, we are satisfied that it is necessary, in order to give business efficacy to those arrangements, to imply a term to the following effect, that if the express definition of Three Month LIBOR ceases to be capable of operation, dividends should be calculated using the reasonable alternative rate to three month USD LIBOR at the date the dividend falls to be calculated. This modifies the term contended for by SC in two respects:
i) First, to make it clear that the identification of the reasonable rate is an objective question, of which the ultimate arbiter is the court, rather than this being an area where SC is permitted to reach its own decision as to an alternative rate which can be challenged only on Braganza grounds (cf. Socimer International Bank Ltd v Standard Bank London Ltd [2008] EWCA Civ 116, [66]).
ii) Second, to allow for the fact that the universe of available alternative reference rates might change over the life of the Preference Shares – e.g. (to take a rather unlikely example) if a robust, transparent and substantial interbank unsecured lending market were to re-establish itself.
- The only suggested alternative to the implication of such a term is the implication of a term requiring the automatic redemption of the Preference Shares as soon as this can lawfully be done. For the reasons we set out at [80] and following below we regard this implied term as wholly untenable.
- We are also satisfied that the term set out at [66] above is so obvious that it goes without saying. As we have stated, it is clear from the relevant contractual terms here that the parties did not intend issues with the availability of a LIBOR rate to prevent the continued performance of this perpetual contract. While Lord Wolfson KC suggested that neither those subscribing to the economic interests in the Preference Shares nor SC would have been willing to run the risk that dividends might at some point be calculated on a basis which would produce a figure materially higher or lower than the LIBOR rate, they had in fact committed themselves to a dividend-calculation mechanism which allowed some scope for changes in the make-up of the rate over time. Each would have required the qualification of "reasonableness" as a sufficient protection of their interests (it being the case that any alternative rate might be higher or lower than 3-month USD LIBOR, or one or other of those things at different times and in different economic conditions).
- We are also satisfied that the term which we have accepted is capable of clear expression, does not contradict any express terms of the contract (which, it is common ground, subject to SC's interpretation argument, do not seek to make provision for the position where the publication of LIBOR ceased altogether) and is reasonable and equitable. While Lord Wolfson KC suggested that the term involved scope for disagreement as to what was reasonable, as Bingham LJ observed in The Didymi at p.116, "different minds may form different conclusions as to what in any situation is fair and reasonable, but that does not import any uncertainty as to the standard to be applied."
- Finally, Lord Wolfson KC suggested that SC's implied term raised the spectre of rival arguments as to which rate to use at each dividend calculation date, given our conclusion at [66] above. We do not accept that submission. While we will not seek to define the criteria relevant to reasonableness exhaustively, we noted that the relevant regulatory bodies had regard to the need:
i) for the alternative reference rate to be based on "a robust underlying market" ([A2]);
ii) to have regard to the liquidity of that underlying market over time, market functioning issues, usefulness to all market participants and the ability to produce and maintain the alternative rates ([A2]);
iii) to be able to test the rate by reference to historical data ([A8], [A9], [A10], [A13] and [A14]); and
iv) to consider resilience to changing market conditions, structures and regulations; placing weight on the diversity of market participants, the stability of their participation and their credit quality over time; how changes in participation could affect the benchmark; the transparency of the benchmark and the need for an ongoing ability to assess the rate's quality ([A18]).
- We also note the widespread use achieved by the Proposed Rate as the alternative to LIBOR (cf. [89]) below).
- The reality is that it has involved many years' work by regulators, analysts and market participants to arrive at the rate which both experts accept is the best available rate, and the Proposed Rate is now a well-established rate used across the financial markets in a variety of financial instruments. The use of the Proposed Rate as a replacement has been endorsed by financial regulators of the major markets in the US and the UK. Further, the benefits of continuity and predictability which will follow from the continued use of the same replacement rate will themselves be matters that can be relied upon to support the contention that this remains the reasonable alternative rate, even if other alternatives become available. Against that background, identifying a viable superior rate will be no mean undertaking, and will become more difficult over time.
- Finally, we note that our conclusion that it is appropriate to imply a term requiring the use of the reasonable alternative rate when an index used to measure payments under a contract ceases to be published derives support from two additional sources.
- The first is the decision of Sir Nigel Teare in Regal Seas Maritime SA v Oldendorff Carriers GmbH & Co KG (The New Hydra) [2021] EWHC 566 (Comm). In that case, the hire rate payable under a time charter for a ship of nearly 180,000 tonnes was to be calculated by taking an average of four Baltic Cape Size Time Charter routes published by the Baltic Exchange over the previous 15 days plus 4% for size adjustment. The adjustment reflected the fact that, when the contract was concluded, the Baltic Capesize Index was produced by reference to a vessel of 172,000 tonnes. However, after the contract was concluded, in July 2015, the size of the assumed index vessel was changed to 180,000 tonnes, with the Baltic Exchange publishing a rate for 172,000 tonne ships derived from the index. Sir Nigel Teare held that it was an implied term of the time charter that if the size of the ship used in the index changed, the size adjustment used to arrive at the hire rate "should be … a reasonable size adjustment" ([33]). He noted at [37]:
"unless one implies the term suggested by the Owners the hire provision would not be capable of being applied in the events which happened after July 2021. This cannot have been what the parties intended."
- At [48], he continued:
"Far from subverting the hire provision and the parties' bargain the implied term ensures that the charterparty continues to operate for the period intended by the parties. It is difficult to see how a rate of hire based upon the new benchmark vessel as reasonably adjusted to reflect any difference in earning capacity can fairly be said to give rise to a windfall, even if, as found by the Tribunal, the Charterers did well out of the agreed size adjustment to the average rate for a 172,000 tonnes vessel."
It is clear that the fact that the reasonable adjustment did not exactly replicate that which was being replaced did not preclude the implication of the term.
- We accept, of course, that The New Hydra was a case in which the Baltic Exchange were continuing to publish an index, albeit for a different sized vessel, with the contractual rate being derived from a reasonable adjustment to that published index. We understood Lord Wolfson KC to accept that if three month USD LIBOR had ceased to be published, but other USD LIBOR tenors had continued to be published from which a three month rate could reasonably be derived, then the use of that derived rate would have been permitted by the contract terms. However, we do not think a line can sensibly be drawn between the use of an alternative rate which can reasonably be derived from one or more different USD LIBOR rates, and the use of the reasonable alternative rate to the relevant USD LIBOR rate when publication of LIBOR rates has ceased altogether.
- Second, we became aware that the-then Leonard Hoffmann QC had been asked to advise the Law Society in 1981 as to the effect on loan contracts requiring interest to be paid by reference to the Bank of England's Minimum Lending Rate ("MLR") of the abolition of that rate. In that advice, which was published in (1981) 78 Law Society Gazette 1029, Mr Hoffmann KC advised as follows:
i) On the abolition of the MLR, there would no longer be any rate which satisfied the contractual definition.
ii) In those circumstances the court would have two alternatives.
iii) It could either "say that the interest rate provision had become inoperative and that accordingly no interest was payable under the contract" (which, in the case of a term loan, would involve frustration of the loan and an obligation immediately to repay the principal).
iv) Or it could "imply a term into the contract to fill the gap" where the appropriate conditions for doing so were satisfied.
v) There was no doubt that without the implication of such a term, the contract would lack business efficacy.
vi) While there was difficulty in deciding precisely what the parties as reasonable men would have said if the difficulty had been drawn to their attention, "the court would make every effort to find an answer rather than accept the consequences of declaring itself defeated".
vii) Either the parties would agree that "the nearest equivalent rate should be substituted", namely the rate which would "as nearly as possible" satisfy the criteria of the MLR (described in the opinion as "the cy-près approach") or imply a term that the court would fix the rate, which "leads straight to the same answer".
- We would note that in the intervening 43 years since that opinion was published, the courts have become even more ready to imply terms to save a contract in circumstances of this kind.
The Funds' Implied Term
- We have set out the Funds' implied term at [37]-[38] above.
- Taking the first stage of the implied term (the automatic redemption of the shares when legal and regulatory conditions allow), in our assessment this fails to satisfy each of the criteria for the implication of a term.
- First, we are not persuaded that the term is necessary to give business efficacy to the contract. The purpose of the contract was for the long-term (and quite possibly perpetual) provision of capital to SC in return for a dividend calculated on a fixed-rate basis for 10 years, and at a floating rate (thereby ensuring the dividend moved in line with the cost of borrowing) thereafter. That purpose can be given effect by the implied term we have found to be established, whereas the Funds' term would bring the provision of capital, and the payment of dividends, to an end. It cannot be said that the contract would lack commercial or practical coherence if it was not automatically determined in the event that LIBOR ceased to be published (and we note that a vast number of financial instruments which used LIBOR as a reference rate have, through a variety of legislative or commercial mechanisms, continued to function using a reasonable alternative rate).
- Second, the term cannot be said to be so obvious that it goes without saying. The term would have involved both parties signing up to a contract in which an event outside the control of SC or those economically interested in the Preference Shares could have brought the arrangement to an end at any time:
i) That would have had obvious implications for SC's ability to make medium and long-term investment and acquisition plans which was one of the uses to which the capital raised could have been put (the very antithesis of what Tier 1 regulatory capital is intended to permit).
ii) For subscribers to the ADSs, it would have involved a risk that the stable income stream which the Preference Shares were intended to provide ceased outside of the very limited circumstances when SC was entitled to redeem them, and a risk of dividends not being declared for a period to enable SC to build up sufficient funds to effect the mandatory redemption. The implied term would have required redemption even if a significant majority of ADS holders (but falling short of the 75% majority necessary for a consensual adjustment) opposed this course.
iii) For both SC and the ADS holders from time-to-time, the implied term would involve the risk of considerable periods of uncertainty while the future of LIBOR was in question (as it was over the extended period set out in the Appendix), with a chilling effect on SC's operations and the marketability of the ADS. If LIBOR ceased to be published during the 10-year fixed rate period, it would also impact SC's operations during the remainder of that period, with the obligation to redeem coming into effect at a fixed point in the future.
- Third, the term is inconsistent with the express terms of the contract, and with the legal controls on the right of redemption which applied when the Preference Shares were issued:
i) The terms on which the Preference Shares were issued provide that "subject to the Articles, provisions of applicable law and to the prior consent of the FSA", SC had the option to redeem the Preference Shares in whole or in part on 30 January 2017 and at 10-year intervals thereafter. This was clearly a right (i) exercisable only by SC; (ii) which was in the nature of an option; and (iii) which arose only at specific intervals. Indeed, the Offering Circular highlighted the risk to potential investors that the Preference Shares and ADSs did "not have a fixed final redemption date and investors will have no right to call for the redemption of the Preference Shares or ADSs."
ii) However, the Funds' implied term would provide for automatic termination (subject to any legal or regulatory restraints) rather than a unilateral option, and one which could arise at any time.
iii) When the Preference Shares were issued in 2006, the "Interim Prudential sourcebook: Banks" issued within the FSA Handbook stated in Sections 5.1(1)(b), 5.14(c) to (e) to the "Tier 1 Capital" chapter that for capital to qualify as Tier 1 capital, (a) preference shares must not "be redeemable at the option of the holder" and (b) preference shares must not have a mandatory redemption date or contain "other provisions which require future redemption of the issue." While Lord Wolfson KC pointed to the mandatory rather than optional nature of the redemption to suggest that the Funds' implied term did not conflict with the first of those requirements, he had no answer to the second.
iv) The statutory restrictions on the redemption of share capital summarised at [11] above require particular caution before implying an automatic obligation of redemption into terms on which share capital is issued.
- Fourth, the proposed term is unclear, in particular:
i) If the FSA/PRA (or their successors) imposed conditions on permitting redemption, as it was plainly permitted to do, it is not clear whether the term would require SC to satisfy those conditions (whatever they might be) or to take reasonable endeavours to do so.
ii) It is not clear if the implied term would require SC when the cessation of LIBOR was in contemplation or had occurred to take reasonable steps to put itself in a position to redeem the shares, e.g. by not paying dividends, selling assets or undertaking a fresh share issue.
iii) During any period in which the obligation to redeem had been triggered by the cessation of LIBOR, but it was not yet legally permissible to redeem, the issue would arise as to what dividend rate should be paid. The Funds offered alternatives of "the last published LIBOR rate … plus 1.51%" (which might have involved a rate wholly out-of-touch with prevailing market conditions, and a windfall to SC or the ADS holders depending on market movements), or the rate which applied during the fixed dividend period (when the parties had clearly agreed a floating rate), or (when pressed) the Proposed Rate. Lord Wolfson KC professed to be indifferent as to which applied, but we do not believe the parties to the bargain embodied in the Preference Shares would have been quite so insouciant, nor regarded the inability to answer that question as raising a separate and distinct question from the implied obligation of redemption.
- Given that the first stage implied term put forward by the Funds fails to satisfy the criteria for the implication of a term for those reasons, it is not necessary to consider the second stage implied term since that would only arise if the first stage implied term was arguable but could not be brought into effect because redemption was unlawful for some reason. Nonetheless, we would note that the acceptance that, in such a scenario, dividends would continue to be paid at an alternative rate (or even at the Proposed Rate) is a further factor undermining the Funds' implied term case.
The position in debt instruments
- In our view, the arguments which have led us to find the implied term at [66] above, and to reject the Funds' implied term, are likely to be similarly persuasive when considering the effect of the cessation of LIBOR on debt instruments which use LIBOR as a reference rate but do not expressly provide for what is to happen if publication of LIBOR ceases. Once again, the use of a floating LIBOR rate as a reference rate in instruments of that kind is essentially a measure of the wholesale cost of borrowing over time. The specific reference to LIBOR is likely, therefore, to be a non-essential term, and the inoperability of the mechanism should not defeat the continuation of the contract.
- Further, any implied term by which the cessation of LIBOR would give rise to an automatic redemption would be at least as, if not more, unworkable in debt instruments, where it would trigger immediate payment of the full amount of the outstanding principal sum without the statutory limitations controlling the redemption of share capital. An implied obligation to redeem the loan on the cessation of LIBOR would have the same accelerating effect as a stipulated event of default in a loan instrument even though:
i) the accelerating event would be entirely outside the control of either party;
ii) events of default in loan contracts are generally carefully defined and clearly set out because of the dramatic consequences they have; and
iii) none of the protective provisions which usually accompany event of default provisions – notice provisions and cure opportunities – would be present.
Is the Proposed Rate the Reasonable Alternative Rate?
- The answer to this issue, on the evidence before us, is straightforward. Dr Sabry and Dr Metz were able to reach agreement on a number of significant issues:
i) Of the available reference rates, the Proposed Rate is the closest to 3-month USD LIBOR.
ii) "Based on an analysis of how well the alternative benchmark rates tracked 3-month USD LIBOR in terms of MAE [Mean Absolute Error] correlation, and volatility, the analysis of the underlying markets of the rates, research of the characteristics of the rates, and the commentary provided by regulators …. [the Proposed Rate] … is the most suitable rate for 3-month USD LIBOR".
iii) The proposed rate is in widespread use: 78% of market issuances in 2023 used SOFR-based reference rates, including instruments issued by a number of international banks. That can be compared with the 78% of floating and variable rate debt issuances indexed to USD LIBOR in 2006, when the Preference Shares were issued.
- We were also referred to a number of issuances which had originally used LIBOR as a reference rate, and which had been amended to use the Proposed Rate in place of LIBOR. We accept Lord Wolfson KC's point that the mechanisms by which they came to do so may in some or all cases have involved something other than party choice – the application of the USD LIBOR Act or similar legislation under other governing laws, or the decisions of calculation agents in contracts which provide such an entity with the right to adopt a replacement rate. However, the fact remains that the Proposed Rate is widely used, including in instruments which might for at least a period be available market alternatives to investing in the ADSs.
- There was considerable disagreement between the experts as to how closely the output of the Proposed Rate would match the output of 3-month USD LIBOR over time and in different economic conditions.
i) Dr Sabry used repo rate data released by the Federal Reserve Bank of New York going back to 1998 ([A10]) to suggest that the Proposed Rate exceeded the 3-month USD LIBOR rate on 83.9% of occasions between February 1998 and June 2023 (based on a rolling five-year median). She told the court that she had performed a similar analysis considering the position on every day during this period and that this showed a similar outcome. The results of this second exercise were not before the court, having been undertaken in response to a second report from Dr Metz which was served relatively shortly before the hearing. We accept, however, that such an outcome would be consistent with the analyses in Dr Sabry's reports.
ii) The data set going back to 1998 used in these analyses was described by Mr David Bowman of the Fed in a publication of 1 July 2019 as being "close enough" to SOFR "to be a reasonable proxy for risk modelling or other purposes." It was also common ground between the experts in the joint memorandum that 90-day SOFR compounded in arrears is a reasonable proxy for 3-month CME Term SOFR (as Dr Metz re-confirmed in his oral evidence).
iii) Dr Metz undertook a different analysis, comparing 3-month Treasury rates and 3-month USD LIBOR going back to 1986, on the basis that 3-month Treasury rates are a reasonable proxy for 3-month Term SOFR. That analysis shows that between 2 January 1986 and 2 January 2019, LIBOR exceeded 3-month Treasury plus the ISDA Spread Adjustment 68.1% of the time.
iv) Dr Metz also performed an analysis using 3-month Treasury Rates for the full life of the Preference Shares (i.e. including the 10-year fixed dividend period) which suggested that a daily dividend calculation (i.e. not simply on the actual dividend calculation days) would have been 4.6% lower than 3-month USD LIBOR. Dr Sabry accepted that if this same analysis was conducted using 90-day SOFR compounded in arrears rather than the 3-month Treasury Rate on an otherwise identical basis, there was a 3.37% difference (which was eliminated if only the floating rate period and contractual dividend dates were considered). However, she took the view that an analysis which was not tied to the contractual period was better conducted by reference to the full available data set going back to 1998 (as to which see (i)).
v) There was a fiercely contested (and not entirely even-tempered) debate between the experts as to the extent to which SOFR and 3-month Treasury rates were comparable. We do not find it necessary to get into intricacies of that debate, beyond noting that we accept that SOFR is properly described as "nearly risk-free", the issue being whether the difference between that and a wholly risk-free 3-month Treasury rate is material for the purposes of the comparison Dr Metz seeks to draw.
vi) On that issue we note that the ARRC, who looked at this issue in some depth with the involvement of numerous market practitioners, identified differences in the way in which US Treasury Rates might perform as against SOFR, particularly in times of stress ([A4] and [A8]). In these circumstances, it is not clear how reliable any comparison going back to 1986 is, while the debate between the experts as to how far economic conditions experienced in the 1980s and 1990s will be replicated in the future is largely an exercise in speculation.
vii) More importantly, we are not persuaded that the comparison going back to 1998 does not provide a fair basis for determining what is the reasonable alternative rate in circumstances in which USD LIBOR is no longer available. That is a lengthy period, involving three major recessions (including the financial crisis and the pandemic). We remind ourselves that the contractual terms signal the parties' readiness to afford some latitude and inherent in any cy-près approach is the possibility of some difference.
viii) Dr Metz suggested that higher USD LIBOR rates in times of stress would have an added value in periods when returns on other investments were likely to be impaired. Dr Sabry accepted that this was in principle the case "all else being equal", but noted that those same market conditions would also increase the risk of non-payment. We accept that this could be a factor (particularly with Preference Shares providing for non-cumulative dividends in which SC's ability to pay dividends might be curtailed in the very conditions when bank lending rates were under extreme stress). The extent to which this factor would counter-balance the benefits of a higher yield in adverse market conditions is essentially a speculative exercise, as is the extent to which such a disadvantage of the Proposed Rate in stressed market conditions might be outweighed by higher payments in other market conditions and/or reduced volatility which might itself appeal to investors.
ix) Finally, Dr Metz rightly made the point that a fixed adjustment like the ISDA Spread Adjustment rather than a dynamic adjustment would not necessarily reflect differences between CME Term SOFR and LIBOR in all economic conditions. However, he did not seek to challenge the conclusion of the FCA (and, we would note, ISDA and the ARRC) that using a dynamic spread adjustment was not operationally feasible, nor did he identify any dynamic spread adjustment which might be used.
- We accept that there are likely to be economic conditions in which the Proposed Rate could be materially less than 3-month USD LIBOR, and vice versa. Indeed, one consequence of using a fixed Spread Adjustment (as the Proposed Rate does) was agreed by the experts to be as follows:
"If the fixed spread adjustment corrects for historical differences between LIBOR and the risk-free rate, it will necessarily understate the differences which would be expected during periods of economic stress while overstating the differences during periods of economic stability".
A significant divergence was seen during a period which both experts reviewed, the commencement of the 2020 pandemic, and we accept that it is possible to conduct an analysis over different periods in which the Proposed Rate performs well, or adversely, against LIBOR, and that the data available for some periods provides a closer proxy for the Proposed Rate than for other periods.
- The ultimate net effect of using the Proposed Rate rather than 3-month USD LIBOR over what will be the life of the Preference Shares can only be a matter of speculation. In terms of present expectations, it is of some interest that there was no move in the price of the ADSs (which are traded) when SC announced its intention to seek a consensual amendment to replace 3-month USD LIBOR with something very close to the Proposed Rate (albeit we accept that this can only bear limited weight). We were also referred to one academic study suggesting that investors have a preference for products using a SOFR rather than LIBOR-linked reference rate because of its reduced volatility (Sven Klingler and Olav Syrstad: "The SOFR Discount", 18 April 2024). Finally, while Dr Metz relied upon the perpetual nature of the investment in the Preference Shares as a reason why the higher LIBOR rates in times of acute market stress were likely to occur, we note that the ARRC regarded the difference between the Proposed Rate and LIBOR as something which was "likely to be smaller the longer the remaining maturity of the loan or security being referenced at the time that LIBOR stops, because any larger errors are more likely to 'average out' over the remaining life of the instrument" ([A14]).
- Nonetheless, we accept that rational market participants may legitimately hold a different view (as Dr Metz evidently does) and the ensuing decades may prove one or other view right. We also accept that different investors may attach different values to the greater propensity of LIBOR to "spike" in times of crisis as against a higher return from the Proposed Rate in calmer times and/or reduced volatility.
- However, even if the Funds had been able to establish that Dr Metz's view was the better view, that would not have the effect that what both experts accept is the best of the available rates is not the reasonable rate for the purposes of the implied term we have found to be established. As we have stated, the contractual terms themselves and the legal principles which fall to be applied in circumstances such as the present admit of the possibility that the reasonable rate may involve a different outcome (financially adverse for one or other party) than if the state of affairs assumed when the contract was concluded had continued to exist. To the extent that there are, or might be, differences in the contractual outcome between the Proposed Rate and 3-month USD LIBOR, they do not come close to placing the Proposed Rate outside the scope of the implied term with all of the adverse and commercially unattractive consequences that would entail.
Conclusion
- For these reasons:
i) We reject SC's case on the interpretation of the phrase "3 month LIBOR in effect".
ii) We find that it was an implied term of the express definition of Three Month LIBOR that if the express definition ceases to be capable of operation, dividends should be calculated using the reasonable alternative rate to three month USD LIBOR at the date the dividend falls to be calculated.
iii) We find that, as matters stand, the Proposed Rate is the reasonable alternative rate.
APPENDIX 1
The US
- In a 2013 report, the Financial Stability Board ("FSB") – an international body that monitors and makes recommendations about the global financial system – identified the decline in wholesale unsecured short-term funding by banks as posing structural risks for unsecured indices such as USD LIBOR. The FSB recommended developing nearly risk-free reference rates, embedding no or only small amounts of credit risk. The Financial Stability Oversight Council, a body established by the US Treasury, expressed a similar view in its 2013 report.
- As a result, in 2014, the Federal Reserve System of the United States ("the Fed") and the Federal Reserve Bank of New York convened the Alternative Reference Rates Committee ("ARRC") to "identify a set of alternative reference interest rates that are more firmly based on transactions from a robust underlying market and that comply with emerging standards". The ARRC initially consisted of representatives from 15 large global investment rate derivative dealers, the Fed, the SEC and the US Treasury, with other stakeholders invited to join as work progressed. The ARRC was charged with identifying a risk-free or nearly risk-free rate that in the consensus view of its members would represent best practice for use in derivative and other contracts. The proposed rate was to be assessed by reference to considerations which included the liquidity of the underlying market over time, market functioning issues, usefulness to all market participants and the ability to produce and maintain the alternative rates.
- Following extensive discussions with representatives of major over-the-counter derivative market participants, international supervisors and central banks, and consideration of six possible rates, the ARRC identified two potential rates as the strongest alternative reference rates in its interim report of May 2016:
i) The Overnight Bank Funding Rate.
ii) Some form of overnight Treasury general collateral repurchase agreement rate.
- One of the rates considered, but discarded, by the ARRC was the rate for Treasury bills of which the ARRC said:
"While … factors imply that Treasury bill or bond rates scored highly against several of the ARRC's main criteria (liquidity, robustness, existence of a term market), there are several criteria that the ARRC felt were not met. Most importantly, Treasury rates respond to safe haven demands. Treasury rates have tended to decline in periods of stress even as other money market rates have risen. Treasury bill rates can also fluctuate due to technical factors and the amount of issuance. ARRC members felt that these properties would make Treasury rates less appealing to most market participants because they would be less effective in hedging private-sector risks and funding costs. Therefore, the ARRC believed it would be difficult to implement a transition from LIBOR to a Treasury benchmark."
- The ARRC also rejected the use of unsecured term lending rates, which were seen as sharing several of the key structural difficulties of LIBOR – limited transactions even in normal times, with substantially fewer transactions in times of stress, and with the sample of firms borrowing at such rates being unstable over time.
- In November 2016, the ARRC announced the formation of an advisory group as part of its efforts to secure market feedback on alternative reference rates.
- On 22 June 2017, the ARRC announced that it had decided on a broad Treasuries repo rate (i.e. an overnight lending rate secured by the deposit of US Treasuries) as its preferred alternative reference rate. It described the rate – referred to as the Secured Overnight Funding Rate or SOFR – as "the most appropriate for wide-spread and long-term adoption as a reference rate", which had the support of a significant majority of the market participants consulted. SOFR was a new rate, which would be published by the Federal Reserve Bank of New York in co-operation with the Office of Financial Research on a daily basis from 3 April 2018.
- The ARRC produced its second report on 5 March 2018. The second report referred to releases of "historical" data for SOFR by the Federal Reserve Bank of New York going back to 2014. This was a reference to the fact that the Federal Reserve Bank of New York had released a range of historical indicative data for the period from August 2014 to March 2018 based on modelled, pre-production estimates of SOFR relying on the same underlying data and methodology as the published rate. In the second report, the ARRC explained that it had recommended an overnight rather than a term rate because of the need for a reference rate "based on a deep and robust underlying market", overnight rates being "judged to be based on the only markets with a deep enough set of transactions to support a rate as heavily used as USD LIBOR." The second report once again explained why the ARRC had not recommended Treasury rates:
"[W]hile Treasury bill or bond rates scored highly against several of the ARRC's main criteria (liquidity, robustness, existence of a term market), there are several criteria that the ARRC felt were not met. Most importantly, Treasury rates are not well correlated with measures of either private-sector financial or nonfinancial corporate borrowing costs … Treasury rates can also move idiosyncratically based on fluctuations in technical supply and demand factors that are not reflective of the true risk-free rate relevant to the market. ARRC members felt that these considerations made Treasury rates less appealing to most market participants because these rates are less effective in hedging private-sector risks and funding costs. Therefore, the ARRC believed it would be difficult to implement a transition from LIBOR to a Treasury benchmark."
In that report, the ARRC anticipated that 92% of assets linked to LIBOR would end by 2025 with only 4% open in 2030 and only 2% open in 2040.
- The report noted that while SOFR was an overnight rate, SOFR could be structured to produce term future rates as was currently done with other overnight rates. The ARRC said that publicly available Treasury repo rate data extended back more than 10 years, and "appear to be reasonable proxies for modelling how a rate like SOFR would have behaved historically." The ARRC set out a phased implementation plan which included creating a forward-looking term rate based on SOFR.
- Shortly after the publication of the second report, on 9 March 2018, the Federal Reserve Bank of New York released data drawn from daily surveys it had conducted since 1998 of primary dealers covering their borrowing activity in Treasury-collateralised repo transactions. Using this data, the Federal Reserve Bank of New York constructed a "time series of the volume-weighted mean rate of the primary dealers' overnight Treasury general collateral repo borrowing activity". While noting differences between this data and SOFR, the Federal Reserve Bank of New York stated, "the survey rate can play an important role in providing insight into how a broad measure of repo market activity would have behaved over a significantly longer time horizon."
- With the ARRC favouring an overnight and nearly-risk free rate, consideration was given in the market to how to address the differences between SOFR and LIBOR (which involved term rates which did reflect credit risk). The International Swaps and Derivatives Association ("ISDA") initiated a market-wide consultation to consider these issues, the results of which were published on 20 December 2018, with a further consultation following, the results of which were published on 18 September 2019. The outcome of these consultations was that the overwhelming majority of respondents "preferred the compounded setting in arrears rate to address the difference in tenors" and "the historical mean/median approach for the spread adjustment to address credit risk and other premia."
- ISDA then undertook a further consultation process to determine the final parameters for the spread and term adjustments, the results of which were published on 15 November 2019. The clear majority of respondents preferred a calculation of a spread adjustment based on a historical median over a five-year lookback period.
- The ARRC had also been looking into the issue of what spread adjustment should be recommended for products referencing USD LIBOR. It produced reports on this issue in January and June 2020. The former report made it clear that the ARRC was only looking at a static spread adjustment, because "dynamic spread adjustments … would need to be based on the same wholesale unsecured funding markets that underpin LIBOR and that have now grown to be so thin". The report looked at the reliability of mean absolute error involved in using SOFR and a static spread for loans with one and five years remaining before maturity and concluded that a static spread and SOFR could "produce results that are as accurate as a potentially dynamic spread." The report appraised ISDA's spread adjustment analysis, which was found to have a sound basis, it being noted that whenever LIBOR spreads had moved away from their long-run values, they had reverted back, generally within a year. The ARRC also noted that the same adjustment seemed to work well across three iterations of term SOFR – a forward-looking term SOFR, a compound average of SOFR in arrears and a compound average of SOFR in advance.
- The report looked at mean absolute errors for different spread methodologies relative to 1- and 3-month LIBOR, using a two-year average, five-year median and a ten-year trimmed mean spread methodologies for the period 1999 to 2019. Mean absolute errors were higher during the financial crisis, the report observing "that if economic conditions remain relatively calm, the spread adjustments to be recommended by the ARRC could have very little associated error". It was also noted that "average annualized errors are also likely to be smaller the longer the remaining maturity of the loan or security being referenced at the time that LIBOR stops, because any larger errors are more likely to 'average out' over the remaining life of the instrument", and that:
"Table 5 (which is based on the full period including the financial crisis) shows that the annualized difference in returns between LIBOR and the potential spread adjustment fallback rates tends to decline as the remaining maturity increases. Because the relationships between LIBOR and other money market rates have historically been fairly stable over long periods of time, any deviations between LIBOR and the potential spread adjusted rates, when they occur, have washed out over time".
The implications of those observations for a perpetual instrument are obvious. The report also noted that a use of a simple average (i.e. mean) would generally have generated larger errors than a median or trimmed mean, with no difference between the two.
- The June 2020 report offered the ARRC's recommendation "for market participants' voluntary use, to produce spread adjustments intended for USD LIBOR contracts that have incorporated the ARRC's recommended hardwired fallback language" (i.e. recommended contractual language to address the discontinuance of LIBOR) and "for legacy USD LIBOR contracts where a spread-adjusted [SOFR] can be selected as a fallback." After a consultation process, the ARRC (like ISDA) recommended "a spread adjustment methodology based on a historical median over a five-year lookback period calculating the difference between USD LIBOR and SOFR".
- On 29 July 2021, the ARRC formally recommended CME Term SOFR rates for the purpose of the transition from USD LIBOR.
- On 15 March 2022, Congress passed the Adjustable Interest Rate (LIBOR) Act as part of the Consolidated Appropriations Act 2022. This contained a series of default rules for legacy contracts which did not provide for the use of a clearly defined and practical replacement benchmark for the period after 30 June 2023 (so-called "tough legacy contracts") and which were subject to US law. For those contracts, the Act made provision for the Board of Governors of the Federal Reserve System to fix replacement rates for those contracts, which it did by regulations issued on 26 January 2023. For non-derivative and non-consumer transactions on USD three-month LIBOR, the Board recommended three-month CME Term SOFR with a fixed spread adjustment of 0.26161% (i.e. the ISDA Spread Adjustment).
- In November 2023, the ARRC published its "Final Reflections on the Transition from LIBOR" which contained a retrospective assessment of its work. The Final Reflections explained that the assessment of possible benchmarks had involved not simply regard for the transaction volumes, but also resilience to changing market conditions, structures and regulations; placing weight on the diversity of market participants, the stability of their participation and their credit quality over time; how changes in participation could affect the benchmark; the transparency of the benchmark and the need for an ongoing ability to assess the rate's quality and robustness. Referring to its earlier conclusion that the available unsecured lending rates suffered from the same structural weaknesses as LIBOR, the Final Reflections noted that the International Organization of Securities Commissions had reached a similar conclusion.
The UK
- In June 2012, the UK government commissioned Martin Wheatley, then Managing Director of the FSA, to establish an independent review into a number of aspects of the setting and usage of LIBOR. That resulted in a report ("the Wheatley Review") published in September 2012 which:
a. recommended the comprehensive reform of LIBOR;
b. identified the very serious risks involved in transition to a new benchmark given the widespread use of LIBOR;
c. concluded that transaction data should be explicitly used to support LIBOR submissions;
d. recommended that the administration of LIBOR be transferred from the BBA to a new administrator;
e. recommended that the BBA immediately cease the compilation and publication of LIBOR for those currencies and tenors where there was insufficient trade data to corroborate the submissions;
f. suggested that market participants should be encouraged to consider and evaluate their use of LIBOR; and
g. recommended that the UK authorities should work closely with international bodies to debate the long-term future of LIBOR and similar benchmarks.
- Administration of LIBOR as handed over to ICE Benchmark Administration Ltd ("IBA"), who began publishing "ICE LIBOR" in 2014 and who were regulated by the FCA. In 2016, the IBA identified issues arising from the substantial decrease in the interbank unsecured loan market in the aftermath of the financial crisis – a trend which the IBA attributed to a number of factors, including an increase in the perceived risk of bank counterparty default, regulatory changes to bank funding requirements, and the increased liquidity available to banks from central banks. That reduction or "thinning out" of the interbank unsecured landing market led the IBA to update the definition of LIBOR to include unsecured lending between banks and non-financial corporations.
- On 27 July 2017, Andrew Bailey, the Chief Executive of the FCA, gave a speech stating that notwithstanding the significant improvements made to LIBOR since April 2013, the absence of an active underlying market raised a serious question about the sustainability of LIBOR benchmarks based on those markets. Mr Bailey stated that the reforms had not realised the hoped-for improvements because "the underlying market that LIBOR seeks to measure – the market for unsecured wholesale term lending to banks – is no longer sufficiently active." He suggested that the absence of active underlying markets "raises a serious question about the sustainability of the LIBOR benchmarks". Market participants were warned not to rely indefinitely on reference rates which did not have active underlying markets to support them. Mr Bailey referred to encouraging panel banks to sustain LIBOR for a period of four to five years, with market practitioners being strongly encouraged to transition from LIBOR in the meantime.
- The dollar volume transactions underpinning the LIBOR submission fell substantially – for the first half of 2017, average daily volumes or 3-month USD LIBOR was estimated to be $500 million. This compared with a market in the same period for 3-month US Treasury bills of $13 billion and much larger markets using one of a number of overnight rates:
a. USD 79 billion on the effective federal funds rate or EFFR published by the Federal Reserve Bank of New York drawn from overnight federal funds transactions;
b. USD 197 billion on the overnight bank funding rate or OBFR which added Eurodollar transactions to the transactions underlying the EFFR; and
c. USD 754 billion on the Secured Overnight Financing Rate or SOFR, a rate based on secured overnight repurchase agreement transactions.
- In September 2018, the FCA and the Prudential Regulation Authority ("PRA") wrote a joint letter to the CEOs of the main supervised banks and insurers asking what they were doing to manage the transition from LIBOR to alternative rates. A similar letter was sent in January 2020.
- In May 2020, the Financial Policy Committee of the Bank of England ("the FPC") warned that the wholesale term funding markets on which LIBOR was based were prone to become volatile and unreliable in stressed periods. In July 2021, the FPC expressed the view that credit sensitive interest rates were not robust or suitable for widespread use as a benchmark and had the potential to reintroduce many of the financial stability risks associated with LIBOR.
- In August 2020, the FPC said that it was essential for firms to end reliance on LIBOR benchmarks before the end of 2021 because, after that, the benchmarks would become unavailable at short notice. The FPC published a similar message in December 2020.
- In March 2021, the FCA and the PRA wrote to market participants encouraging them to move away from USD LIBOR to SOFR, and the Bank of England and the FCA issued a joint statement drawing attention to the end dates for the panel bank LIBOR submissions.
- On 5 March 2021, the FCA announced that the IBA would stop publishing 24 currency and tenor LIBOR settings after December 2021. Some GBP and JPY settings would continue to be produced on a synthetic basis, with USD settings continuing until 30 June 2023. The FCA observed of the synthetic LIBOR rates that they "will no longer be representative of the underlying market and economic reality the setting is intended to measure."
- In June 2021, the Bank of England and the FCA issued a joint statement supporting the US-led initiative to shift liquidity from USD LIBOR to SOFR from 26 July 2021.
- On 31 December 2021, publication of 24 of the-then remaining 35 LIBOR settings (currency and tenor combinations) ceased. Three JPY LIBOR settings were continued on a synthetic basis, along with 1-, 3- and 6-month sterling LIBOR and there would continue to be panel LIBOR rates for the overnight 1-, 3-, 6- and 12-month USD settings. However, use of those USD LIBOR settings in new contracts was restricted.
- In June 2022, the FCA carried out a consultation on ceasing synthetic sterling LIBOR and sought information on whether to continue to require USD LIBOR to be published on a synthetic basis for a limited period after the end of the US dollar LIBOR panel at the end of 2023. The FCA stated the proposed synthetic USD LIBOR rate was expected to comprise a forward-looking term rate of SOFR "plus the respective ISDA fixed spread adjustment".
- In November 2022, the FCA carried out a further consultation on whether it should require publication of 1-, 3- and 6-month USD LIBOR settings on a synthetic basis until the end of September 2024, "as well as on the appropriate methodology of constructing such synthetic US dollar LIBOR settings and what use of them should be permitted." The FCA reported that it had decided it "should follow the same approach for US dollar LIBOR settings as for sterling and yen LIBOR settings", using the CME Term SOFR rate which it described as "robust and transparent." In a paragraph which made a number of appearances in SC's filings, the FCA stated:
"As we have said previously, while in our view synthetic LIBOR settings are a fair and reasonable approximation of what LIBOR might have been had it continued to exist, they are not representative of the markets that the original LIBOR settings were intended to measure."
- The USD LIBOR bank panel ended on 30 June 2023.
- Through the combination of Regulation (EU) 2016/1011 of the European Parliament and Council of 8 June 2016, s.15(1) of the Financial Services Act 2021 and the Critical Benchmarks (References and Administrators' Liability) Act 2021, the FCA was given power to set the method by which a benchmark in a contract or any other arrangements would be determined.
- On 14 March 2023, the FCA exercised this power, requiring the IBA to publish synthetic rates for 1-, 3- and 6-month USD LIBOR to support a transition from LIBOR effective from 00.01 on 1 July 2023.
- On 1 July 2023, the FCA used its regulatory powers to require the IBA to continue to produce synthetic LIBOR for 1-, 3- and 6-month USD LIBOR rates "based on the relevant CME Term SOFR Reference Rate plus the respective ISDA fixed spread adjustment", warning that the synthetic rates "will no longer be representative of the underlying market that LIBOR was intended to measure". The FCA stated that it was satisfied that this was a "fair and reasonable approximation of the value panel-bank LIBOR would have had".
- On 17 May 2024, the FCA published a further notice extending the period for which synthetic USD LIBOR would be published to the end of September 2024.
- At the end of September 2024, synthetic USD LIBOR ceased to be published.
