Confirmations for interest rate derivatives typically incorporate the 2006 ISDA Definitions. Other definitions booklets for other derivative products that reference IBORs (e.g. equity, FX and commodity) also exist. All of these booklets, to a greater or lesser extent, the consequences of the unavailability of an IBOR but, with some exceptions, they generally only work effectively where the unavailability is temporary. They are not therefore always suitable where the IBOR has been permanently discontinued or, for that matter, where any required regulatory approval for administration, contribution or use of the IBOR does not or ceases to exist or is suspended.
In order to avoid situations whereby the performance of a derivative would be compromised, ISDA has developed a Benchmarks Supplement (including various product-specific Annexes), with a primary aim of addressing the requirements of the EU Benchmarks Regulation. The Benchmarks Supplement includes various fall-backs to deal with permanent IBOR discontinuance or benchmark regulatory non-compliance (where these events are not already addressed in the relevant Definitions booklet).
ISDA is carrying out a separate work stream, which is ongoing, to supplement the 2006 Definitions. When the relevant supplement is published, this will create so-called “priority fall-backs” in relation to certain designated IBORs that will be triggered on the occurrence of an “index cessation event”. This is designed to ensure that, where a specific IBOR is intended to transition in an orderly manner to a specified RFR, (i) the transition occurs as intended: and (ii) in theory, all transactions that incorporate the 2006 Definitions will transition at the same time and in the same manner. The Benchmarks Supplement contractually provides that these priority fallbacks will apply to transactions that incorporate the 2006 Definitions instead of the more generic fallbacks set out in the Benchmarks Supplement, which are described in more detail below.
The Benchmarks Supplement and the 2006 Definitions (as supplemented) are, as noted, intended to assist users of benchmarks to comply with their obligations under the EU Benchmarks Regulation and to provide for the consequences of a permanent cessation of (or of regulatory non-compliance with respect to) a benchmark, such as an IBOR, in the context of transactions that incorporate one or more of the 2006 Definitions, the 2002 ISDA Equity Derivatives Definitions, the 1998 FX and Currency Option Definitions and the 2005 ISDA Commodity Definitions.
The Benchmarks Supplement was published in final form on 19 September 2018. Legacy and/or new trades that incorporate the Benchmarks Supplement (either by way of (i) the ISDA 2018 Benchmarks Supplement Protocol published on 10 December 2018 – itself available on the protocols section of the ISDA website and on ISDA Amend; or (ii) bilateral negotiation) will automatically benefit from its fall-back provisions.
The Benchmarks Supplement provides for a waterfall of fall-backs that could apply on a permanent cessation of (or regulatory non-compliance with respect to) an IBOR, which may include:
- The application of a priority fall-back if one is included in the relevant ISDA definitions booklet for that benchmark
- The parties agreeing on the modifications to be made to the affected transaction to account for the permanent cessation of that benchmark
- The parties pre-agreeing that a certain replacement benchmark will apply
- The parties agreeing that a replacement benchmark that has been nominated by a relevant central bank or supervisory body will apply; or
- The selection by the Calculation Agent (subject to various controls, checks and balances) of an alternative benchmark.
Save for the priority fall-backs, which operate to the exclusion of the other fall-backs, these fall-backs operate in parallel, rather than sequentially, and, in the event that none of them provides a workable solution, the ultimate remedy is no fault (i.e. mid-market) termination. If more than one fall-back produces a viable solution, a pre-determined hierarchy operates to determine the actual outcome.
Since IBORs and RFRs are fundamentally different economic metrics, the straight substitution of an IBOR with an RFR by itself will drastically change the risk profile and economic reality of the relevant transaction. To cater for this, the Benchmarks Supplement provides that, where a fall-back other than a priority fall-back is applied with respect to a transaction, either an adjustment payment (i.e. a one-off payment between the parties) is to be paid or an adjustment spread (i.e. a spread added to or subtracted from the floating rate leg) is to be built in to the transaction. The priority fall-backs include pre-determined mechanisms for determining how an adjustment spread should be calculated, rather than leaving it to the discretion of the Calculation Agent for a particular transaction, in order to avoid changes to the mark-to-market valuation of that transaction.
Note that, in the absence of an adjustment spread/payment, the mark-to-market for the fixed rate payer (typically an end-user/borrower) under an interest rate swap is likely to decrease (because RFRs are historically lower than the IBORs they are intended to replace, so the fixed rate payer will receive less under the contract going forward). This could have significant P&L, accounting and cash-flow consequences for such entities. Note, additionally, that an adjustment spread may be preferable to a one-off payment for liquidity reasons.
The ‘trouble’ with the Benchmarks Supplement is that it is a stand-alone solution adopted for the derivatives markets. If – in relation to, say, a loan that is hedged by an interest rate swap – the parties incorporate into the former an LMA-driven IBOR-replacement solution and incorporate into the latter the Benchmarks Supplement, it is almost inevitable that IBOR discontinuance will give rise to economic mismatch i.e. basis. To the extent that basis risk of this sort existed in the transaction before IBOR discontinuance, this may not be an issue. But in more cash-flow critical structures, the problem will need to be addressed. In addition, the Benchmarks Supplement does not address the mechanical issues to which IBOR discontinuance or replacement gives rise. This will take place through a rewrite of the relevant Definitions booklets expected to be completed during 2020
A further issue for the OTC derivatives market is whether the replacement of IBOR in legacy OTC derivative contracts will constitute a material change to such contracts i.e. a life-cycle event for the purposes of EMIR and Dodd-Frank, which in turn may require such contracts to be margined and/or centrally cleared. After much to-ing and fro-ing and pronouncements/intervention from the FCA, ESMA, BCBS/IOSCO, the ECB, the ARRC and the US regulatory authorities, the position has been now clarified. In both the EU and the US, an amendment to a contract for the sole purpose of introducing IBOR discontinuance fall-back language will not, of itself, constitute a life-cycle event for the purposes of margining or clearing obligations (although note that any subsequent changing of the reference rate pursuant to the operation of a fall-back will do so). See, in this regard, ESMA’s December 2019 statement and the CFTC’s December 2019 announcement.
In this context, in January 2020, ISDA lobbied the ESAs to try to get them to modify a proposed amendment to EMIR that seeks to clarify that clearing and margin requirements (for uncleared trades) should not apply if contracts are ‘novated’ to introduce fall-back provisions in respect of interest rate benchmarks, as required by the EU Benchmarks Regulation.
The modification would seek to have the clarification extended to all benchmarks (equity, credit, commodity, fx etc.) not just interest rate indices.
Note additionally in this regard:
– an anti-avoidance provision is proposed whereby any attempt to improperly avoid margining or clearing obligations will remain subject to enforcement proceedings by relevant regulatory authorities;
– it is widely acknowledged that the word ‘novated’ is neither accurate nor helpful. ISDA has previously lobbied the ESAs on this without much luck; but the policy intent is clear – the introduction of fall-back provisions in legacy contracts should not of itself trigger clearing or margin requirements in uncleared trades.
A separate and final consideration is the extent to which IBOR replacement will necessitate accounting, regulatory capital and other adjustments in relation to affected transactions, especially where IBOR discontinuance gives rise to basis of the sort discussed above.
At the request of the Financial Stability Board’s Official Sector Steering Group, ISDA has led initiatives to establish appropriate fall-backs (such as RFRs) for certain IBORs to address the permanent discontinuation of such benchmarks. Specifically, ISDA’s working groups aim to:
- Identify fall-back triggers
- Identify alternative rates and other fall-back mechanisms
- Identify strategies for fall-back implementation
- Amend the 2006 ISDA Definitions and other definitions booklets, allowing fall-backs to be incorporated into new trades, and
- Create a protocol to incorporate fall-backs into legacy trades.
Much of this work has been completed.
In addition, ISDA, as part of a group of financial market associations, conducted a worldwide review in order to gauge readiness for the transition.
In June 2018, ISDA published a transition report, highlighting the issues raised by market participants, the most significant of which were:
- The lack of standardised documentation making the transition challenging for cash market products
- Possible basis risk arising from the lack of a standardised fall-back mechanism between financial products. Specifically, alternate fall-back rates or timings for fall-back may result in unintended exposure; and
- The risk that synthetic LIBOR continues to be submitted by dealers.
On 12 July 2018, ISDA launched a consultation on technical issues related to benchmark fall-backs for derivative contracts to address some of the items highlighted in the transition report. The ISDA consultation sought input from all (and not just derivatives) market participants regarding various approaches to term and spread adjustments that will need to be made to fall-back RFRs after the transition becomes effective to more fully align the economics of the new RFRs to the defunct IBORs. Those issues needed to be addressed because the fall-back RFRs are overnight and risk-free (or nearly risk-free), whereas the relevant IBORs have term structures and incorporate a bank credit risk premium and a variety of other factors (e.g. liquidity, fluctuations in supply and demand).
These issues and the solutions proposed by ISDA to market participants in the ISDA Consultation were:
SOFR and other RFRs are overnight rates whereas IBORs have different rates for specific tenors. Term rates are generally preferred by corporates because it enables them to quantify payments in advance. In order to address this issue, the ISDA Consultation described and requested feedback on four possible approaches to term adjustments listed below:
- Spot Overnight Rate – The fall-back rate would be the RFR that sets on the date that is one or two business days (depending on the relevant IBOR) prior to the beginning of the relevant IBOR tenor
- Convexity-Adjusted Overnight Rate – The fall-back rate would be the spot overnight rate but it would be modified to account for the difference between flat overnight interest at the spot overnight rate versus the realised rate of interest that would be delivered by daily compounding of the RFR over the IBOR’s term. This is achieved by using an approximation in which “today’s” overnight RFR is assumed to hold constant at “today’s” value on each day during the relevant IBOR tenor
- Compounded Setting in Arrears Rate – The fall-back would be the RFR observed by looking backwards over the relevant IBOR tenor and compounded daily during that period. One issue identified in connection with this adjustment is that the rate would not be known until the end of the relevant period. This would not enable market participants to know the rate, and therefore the amount of accrued interest, in advance
- Compounded Setting in Advance Rate – This approach would be similar to the compounded setting in arrears rate but while the observation period would be equal in length to the IBOR tenor, it would end immediately prior to the start of the relevant IBOR tenor so that the rate would be available at the beginning of that period.
Credit Risk Spread Adjustments
IBORs take into account bank credit risk while RFRs do not. In order to avoid significant value transfer to occur because of this difference on the day the fall-back to the relevant RFR is activated, the ISDA Consultation proposed the following three possible approaches to incorporate a spread adjustment to RFRs:
- Forward Approach – The spread adjustment would be based on observed market prices for the forward spread between the relevant IBOR (to the extent such forward curves exist at the relevant time) and the adjusted RFR in the relevant tenor at the time the fall-back is triggered. This would entail comparing the two curves (relevant IBOR and fall-back RFR) and the spread between those two curves would apply. The spread would be static and would therefore not change at all. That being said, the spread in reality is not a constant number but will vary based on the differences between the two curves going out for the remaining tenor of the relevant contract
- Historical Mean/Median Approach – The spread adjustment would be based on the mean or median spot spread between the IBOR and the adjusted RFR calculated over a significant, static look-back period (e.g. 5 years, 10 years) prior to the relevant announcement or publication triggering the fall-back provisions. This approach also provides for a one year transitional period (subject to linear interpolation) and would become finally fixed at the end of the transitional period
- Spot-Spread Approach – The spread adjustment could be based on the spot spread between the IBOR and the adjusted RFR on the day preceding the relevant announcement or publication triggering the fall-back provisions. This is the same as the second (historical mean/median) approach but applied over a shorter period of time.
ISDA consulted on the following IBORs: GBP LIBOR, CHF LIBOR, JPY LIBOR, TIBOR, Euroyen TIBOR and BBSW. In addition it sought preliminary feedback in relation to USD LIBOR, EUR LIBOR and EURIBOR. The final report, published on 20 December 2018, highlighted that the overwhelming majority of respondents preferred the ‘compounded setting in arrears rate’ for the adjusted RFR and a significant majority across different types of market participants preferred the ‘historical mean/median approach’ for the spread adjustment. It also noted that the majority of respondents preferred to use the same adjusted RFR and spread adjustment across all benchmarks covered by the consultation and potentially other benchmarks (including those on which only preliminary feed-back was sought).
In May 2019, ISDA launched two further consultations on benchmark fallbacks – one covering adjustments that would apply to fallback rates in the event certain IBORs were permanently discontinued, and another, with encouragement before and after from the FSB and the OSSG, relating to pre-cessation issues for LIBOR and certain other IBORs.
The first of these consultations set out options for adjustments that will apply to the relevant RFRs if fallbacks are triggered for derivatives referencing US dollar LIBOR, Hong Kong’s HIBOR and Canada’s CDOR. Feedback was also sought on a proposed fallback for Singapore’s SOR following a permanent cessation of US dollar LIBOR, given US dollar LIBOR is currently used as an input to calculate the Singapore rate.
When the results of the first consultation were announced in September 2019, they confirmed preliminary findings 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’ to address the difference in risk premia. One result of this was that the LIBOR-SONIA basis narrowed sharply as the market reacted to the news.
The results of the second consultation, on pre-cessation, published in October 2019, were inconclusive. As a result, in December 2019, and responding to its members concerns, ISDA sought clarity on two issues: (i) that following a statement of LIBOR unrepresentativeness, the rate would continue to be published for a short time only i.e. months as opposed to years; and (ii) that CCPs (a) would change their rules so as to permit them to transition their portfolios to RFRs following the relevant statement; and (b) would act on those rules at the relevant time. ISDA subsequently (in January 2020) published two letters addressed to it, one from the FCA and one from the IBA, each confirming for a variety of reasons that, if LIBOR became unrepresentative, it would continue to be published for a short period thereafter at most – and not years as some had suggested/feared. It seems clear, therefore, that, following an announcement or other indicator(s) of unrepresentativeness, LIBOR (even zombie LIBOR) would cease to be published fairly quickly. In addition, LCH has indicated that it is considering aligning its rule book to ISDA pre-cessation triggers (wherever and whenever the dust settles on that) so that the cleared and uncleared markets move in tandem on a pre-cessation. Based on those assurances, ISDA announced in February 2020 that it is re-consulting its members to seek consensus as to whether a pre-cessation trigger should be included in new and/or legacy contracts.
At the same time as the two further consultations, ISDA launched a third consultation to finalise the methodologies for the relevant adjustments. The results of this consultation, published in November 2019, showed that a majority of participants preferred a historical median approach over a five-year lookback period. A majority also preferred not to include a transitional period in the spread adjustment calculation, not to exclude outliers, and not to exclude any negative spreads. For the compounded setting in arrears rate, a clear majority favoured a two-banking-day backward shift adjustment for operational and payment purposes. To assist market participants, ISDA published FAQs on IBOR Fallback Rate Adjustments on its website in January 2020 (updated February 2020).
In December 2019, ISDA launched a fourth, supplemental consultation on the spread and term adjustments that would apply to fallbacks for derivatives referencing euro LIBOR and EURIBOR in the event those benchmarks are permanently discontinued.
Based on the results, ISDA plans to make relevant adjustments to the 2006 ISDA Definitions to incorporate fallbacks for new IBOR trades. A protocol will also be published to enable market participants to include fallbacks within legacy IBOR contracts if they choose to. Both the amended Definitions and the protocol are expected to be finalised in Q1 2020, with implementation later this year.
It is worth noting that the spread adjustment discussed above will inevitably not be completely economically neutral and will result in value transfers. However, the only real way to avoid this is to use actual market quotations for basis swaps between the relevant IBOR and compounded RFR over the remaining life of the trade. The issue with that is those quotations will either not be available or the bid-offer spreads will be extremely wide because, at the time of sourcing the quotes, the IBOR will have either been discontinued or been declared unrepresentative. ‘Trading out’ in this manner will inevitably have to done on a trade-by-trade basis, which will be burdensome.
Having said the above, looking backwards over a relatively long period to determine a historic average for the relevant basis is a decent compromise, particularly as the basis between IBORs and the corresponding RFRs is relatively stable (except for around 2008, but that period is not included in ISDA’s proposed lookback period and is very much an anomaly). So, while the period is somewhat arbitrary, it ought to be representative of what the basis would have been going forward and it should therefore keep value transfers low. It is also very easy to implement as a third-party (i.e. Bloomberg) will perform the spread calculations, which should therefore leave less room for disputes.
A separate issue is that the fallback methodologies adopted by the derivatives market, on the one hand, and the cash markets, on the other, are unlikely to be aligned, leading to bifurcated economic outcomes and problems for hedging arrangements.
Frequently Asked Questions
Q. Can’t I just substitute an RFR for IBOR in my swap documentation?
A. No – the metrics are different and without an adjustment (e.g. to the spread) to compensate for this, the transaction economics will be fundamentally altered. In addition, floating amounts will accrue differently, payment mechanics will alter and systems will need to be modified accordingly.
Q. Is a Protocol or similar solution to the problem contemplated?
A. Yes (and it has now been published). This enables parties (through ISDA Amend or via the ISDA web-site) to incorporate the Benchmarks Supplement into designated transactions. Parties are of course free to repaper on a transaction by transaction basis. Relevant definitions booklets will also be amended in due course to cater for mechanical issues.
Q. How does the Benchmarks Supplement work?
A. Click on the link: ISDA Benchmarks Supplement – FAQs
Q. How does the Protocol work?
Q. What is the position with cleared derivatives?
A. The primary concern is market fragmentation i.e. that clearing houses will, under their rules, adopt different fallback methodologies for cleared transactions than those adopted by the OTC market. However, some comfort has recently (January 2020) been given in this regard by LCH, which has indicated that it is considering aligning its rule book to ISDA pre-cessation triggers (wherever and whenever the dust settles on that). The task is huge, however. According to an FCA report, there were, as of August 2019, over £25 trillion of cleared sterling LIBOR-pegged derivatives outstanding, with half having maturity dates past the planned cessation of LIBOR in late 2021.
Q. What about exchange-trade derivatives?
A. We expect that the exchanges will, like the clearing houses, adopt parallel fallback methodologies to the OTC market, although we have seen little evidence of this so far.
Q. What about swaptions?
A. The issues are complex. See chapter 4.1.4 of the ECB’s August 2019 paper on transition from EONIA to ESTER for a good discussion of the topic. In the cleared space, LCH has said that it will not be developing an industry-wide method for compensating swaptions holders when the benchmark for interest rate swaps shifts to RFRs but rather that compensation will be addressed on a transaction-by-transaction counterparty-by-counterparty basis. ISDA has produced (in late 2019) two papers on the issues (both in draft and ongoing). In February 2020, the BofE WGRFR published a survey of 15 dealers in non-linear sterling interest rate derivatives to understand the preferred approach of market participants for the trading of interbank SONIA swaptions and caps and floors.
Q. What about currency swaps?
A. These are in principle no different to interest rate swaps – there are just two legs to consider. The first inter-bank cross-currency swap with both legs pegged to an RFR was traded in November 2019.
Q. What about IBORs in CSA/CSD documentation?
A. On 14 February 2020, ISDA published the ISDA Collateral Agreement Interest Rate Definitions (available on www.ISDA.org). These enable parties to include standardised definitions relating to overnight interest rates in ISDA collateral agreements.
Check list of things to consider
- What is my intended RFR?
- When do I want the change to take effect? Note in this regard that teh BofE and the FCA in December 2019 issued a statement to the market encouraging sterling LIBOR-based interest rate swaps to more to SONIA from Spring 2020
- Is there to be a spread adjustment or one-off payment?
- How do I intend to amend my documentation? If through the ISDA Benchmarks Supplement, then how? By Protocol or bilaterally? Do I adhere or repaper with respect to new transactions only or new and legacy? If the latter, does that constitute a life-cycle event?
- Are there regulatory capital and/or compliance implications?
- Are there implications for payment mechanics, mark to market etc.?
- Are there accounting and tax implications as well as implications for P&L generally?
- Is the transaction a hedge for another instrument? If so: (a) how is IBOR disappearance being dealt with in that instrument; (b) do I need to follow suit in the derivative transaction; (c) do I need to carve the transaction out of the Benchmarks Supplement?
- Can I compress my portfolio to alleviate some of the problem?