The loan markets have accepted that the FCA’s pronouncements on the future of LIBOR have signalled the likely demise of the widely-used benchmark, alongside other IBORs, and the need to develop suitable alternatives.
Some commentary in the loan market has looked at whether a reformed LIBOR with a strengthened methodology (and relying on voluntary submissions, albeit in what is likely to be a continually contracting market) might be able to fill the gap, dubbed “LIBOR+”. Alternatively, a rate derived from the last published rate may survive (variously known as “Zombie LIBOR”, “Synthetic LIBOR” or “Static LIBOR”). However there is no certainty on any of this and the message from central banks, global regulators and the Financial Stability Board is that it is not sensible or appropriate for the loan markets to rely on LIBOR, in any form, after 2021.
Separately, in relation to EURIBOR, the LMA has expressed concerns as to how a hybrid methodology would sit alongside existing loan market definitions and calculation mechanisms (LMA letter, dated 15 May 2018). In the case of both legacy transactions that may need repapering and also the origination of new loans, lenders opening discussions on pricing will need to be mindful of their regulatory obligations in terms of transparency and fairness as well as competition law liability when discussing pricing adjustments.
The current focus of the loan markets is on identifying and modifying the methodology relating to RFRs for use as RFR-based alternatives. However there are a number of reasons why RFRs in their current form do not work for loans:
- RFRs do not have a term element (although ICE Benchmark is looking to publish forward-looking term rates based on derivatives that reference RFRs, starting with SONIA)
- As explained in What is LIBOR, LIBOR is a forward-looking rate, with seven tenors, whereas RFRs are, for the time being at least, backward-looking rates that relate only to overnight periods. By way of illustration, LIBOR is set at the beginning of the interest period at 11am London time. That determines the amount of interest payable at the end of the relevant interest period. Therefore, as borrower, you have predictability, knowing what your interest payment will be. Aside from the uncertainty in not knowing your interest burden until the point of payment, loan systems would need to be entirely rebuilt to calculate interest due on the relevant interest payment date on the basis of backward-looking, overnight rates
- As a means of calculating the cost of interbank lending, LIBOR already prices in credit risk, whereas RFRs are near risk-free
- LIBOR includes a premium on long-dated money, while current RFRs do not. That is to say, RFRs do not compensate lenders for making money available long-term. This leaves a credit and liquidity spread (or ‘pricing gap’) between LIBOR rates and RFRs (on an overnight indexed swap basis)
- Some, but not all, RFRs are secured rates. Others are unsecured. LIBOR is an unsecured rate, which may cause issues in cross-currency swaps, see Derivatives
- Currency issues are magnified, particularly with multi-currency facilities, as LIBOR methodologies are the same for all currencies whereas this is not necessarily true for RFRs. For example, the United States dollar RFR, SOFR, is a secured overnight rate whereas SONIA is unsecured. Currently, in a multicurrency facility, the interest rate in respect of the dollar and sterling facilities would both be calculated using a LIBOR benchmark, albeit on USD LIBOR and GBP LIBOR respectively
- RFRs are likely to have different local methodologies and will be published at local times. This is not the case with IBORs
- Not all currencies have RFRs yet and the transition timetables to RFRs also differ globally (although efforts are under way to coordinate across regulators, central banks and rate administrators)
- Existing documents are built around features and conventions of LIBOR as a forward-looking term benchmark (e.g. quotation days, interest periods, payment dates, currency options etc.)
- Selected benchmarks will need to align with the derivatives markets as any mismatch will create basis risk (Derivatives) and affect liquidity where loans are hedged.
The effect of IBOR discontinuance will vary depending on whether one has:
- An existing loan agreement maturing or with payment obligations falling after 2021 when an IBOR may no longer be published, that includes “fall-back” provisions, which may vary from deal to deal;
- An existing loan agreement that contains no “fall-back” provisions; or
- A new loan agreement for a loan maturing or with payment obligations after 2021 where you have an opportunity to draft provisions to deal with any discontinuance of IBORs.
It should also be noted that there is no protocol solution for loan documents (as may be the case in other markets, see Derivatives); legacy contracts will need to be amended on a case by case basis. Regardless of where the market settles on alternatives to LIBOR, it is inevitable that a significant majority of existing loan agreements will need amendment. Additionally, new deals that are documented from this point will likely include future amendment options. For some, larger syndicated facilities this will require the engagement of dozens if not hundreds of lenders. In any amendment scenario there is the possibility of an unfavourable outcome for one or more parties and/or a failure to reach a consensus. This may activate fall-back provisions that are equally undesirable and may, if widespread, create instability and market disruption.
Working group engagement
Proactivity in the loan market is largely limited to engagement with the regulatory bodies. Counterparties have consistently stressed that solutions must be “market led”. The LMA and the Association of Corporate Treasurers (ACT) are participating in the BoE Working Group on Sterling RFRs. The focus was initially on reforming SONIA. LMA and ACT participation is in sub-groups on benchmark transition issues in syndicated loan markets and on term-rates.
The LMA has acknowledged that the initial work by the BoE was led by the derivatives market and there was comparatively late recognition from the BoE (and others) about the unique issues that are, or will be, encountered by the cash markets. The LMA and the ACT have since actively urged borrowers and lenders alike to engage with the process directly and via the working group channels.
The loan market transition sub-group is focused on raising market awareness of the issues, developing suitable documentation for transition and also new rates, developing contingency plans for legacy contracts, promoting the adoption of SONIA in the loan markets, coordinating transition of non-GBP LIBOR and mitigating challenges from a switch to SONIA.
The sterling RFR working group has found a use case for term SONIA rates in the loan markets and is considering conventions for using SONIA compounded over various interest periods.
While the UK focus is, understandably, on SONIA, market participants and advisers also need to be aware of (and engage where necessary) in the timetables and outcomes for the development of RFRs for other leading currencies (notably USD, HKD, JPY, CHF, EUR, AUD, SGD etc.). In particular the European landscape is complicated by the EU Benchmarks Regulation, which from January 2020 will require benchmarks that are used in certain types of financial contracts (which excludes corporate loans) to be authorised for use within the EU. For more on SOFR, and ESTER (for USD and the euro respectively), see here.
The FCA and FSB have recognised the need for RFR-derived term rates but it may not be practical (or even desirable) to create term RFRs in some currencies where markets are insufficiently liquid. However, regulated entities may need to demonstrate why they might determine it was or is appropriate to enter into a relevant IBOR loan at the time of execution. In all instances the related infrastructure to deal with term RFRs both on the lender and borrower side, will need to be developed before they can be rolled out to the loan markets.
Borrower side views
Anecdotal evidence suggests borrower engagement with the issue is mixed. Since there are presently no RFRs suitable for the loan markets, the view as of today is “wait and see”. That being said, we understand that larger corporate entities (FTSE 20 at least) are talking to lenders and are engaged in the working groups. For borrowers outside of this group, engagement falls off dramatically. The view may be that it is “tomorrow’s problem” and that the lawyers and lenders will sort it out and present solutions to the market in due course. However, all parties should be encouraged to consider the issues relevant to their business now, and not defer in the hope that suitable market solutions will be developed.
Borrowers are likely to be anxious about anything that might encourage banks to revise pricing upwards and particularly to pass on market-related costs that have, since the demise of the mandatory costs schedule, previously been excluded. Accommodating any pricing adjustment in the margin is less transparent for borrowers since a margin typically reflects credit risk and not funding cost (and “backup” rates may reflect higher cost of funding).
Changes based on pre-emptive triggers may be unwelcome as borrowers may assume a lender is more likely to want to transition to new rates (or activate a fall-back position) where LIBOR is no longer as attractive. There may also be issues with transitioning to new rates or methodologies where the infrastructure and conventions are not fully developed for the loan markets.
Equally, the necessity to re-open legacy contracts in particular may encourage refinancing and buy-backs, which may, depending on where things are in the business cycle (or any change in credit quality), present opportunities to renegotiate a better deal elsewhere. Refinancing may be preferable if lenders are going to try to reprice, especially if debt matures in the near term. In each case, there’s no guarantee any party will get a more favourable outcome.
Borrowers will want to maintain the forward-looking and term features of current benchmarks that allow for certainty of funding costs and therefore cash flow planning. For example, calculating interest costs in advance allows for a more informed repayment or refinancing strategy.
A backward-looking or overnight rate may require cash retention to accommodate rate movements during the period between interest payment dates.
Borrowers should consider the wider impact of new benchmark rates on taxation, intercompany debt and cross-border accounting.
Borrowers normally pay amendment costs, but will be reluctant to pay for changes that may increase the cost of their loans. Even if the amendment is neutral on costs, it represents a departure from the norm since the change is lender-driven. Some borrowers may look to repay or prepay if agreement on IBOR replacement be cannot achieved (assuming such rights exist).
The LMA and the ACT have produced a joint guide to the proposed transition away from LIBOR-based benchmarks. The sterling RFR working group has also published a paper on New and Legacy Loan Transactions referencing LIBOR. It is anticipated that a note on best practice for referencing SONIA in the loan markets will follow.
Current LIBOR loans
LIBOR is used extensively as the interest rate benchmark for loans. Interest is typically comprised of a forward-looking term rate (e.g. 1 month, 3 month or 6 month LIBOR) plus a margin (being a fixed spread that reflects the creditworthiness of the borrower).
Interest periods are selected by a borrower. LIBOR is set at the beginning of an interest period and interest, being LIBOR plus the margin (the margin being fixed at the time of entering into the loan), is paid on an interest payment date at the end of the relevant interest period.
Repayments of principal then align with interest payment dates otherwise the borrower must pay break costs to compensate the lender for interest that would have been payable had the payment not been made inside the interest period. This reflects a lender’s matched funding obligations (as relevant) to the interbank lending market, which funds many commercial loans.
Legacy fall-back options
Historically, documents that are based on LMA provisions that pre-date the Screen Rate Rider first published in May 2018, and which may extend to bilateral loans, typically include two fall-back options (“Legacy Fall-back Options“). The Legacy Fall-back Options are suggested by the LMA templates but, they are open to negotiation and modification so may vary from deal to deal. As drafted, they assume parties will use the relevant Screen Rate as the benchmark rate, all the while it is available. Even where the Screen Rate is available, where the interest rate on a facility is hedged, the swap may have moved to a different rate, giving rise to basis risk (e.g. a negative hedge position as a result of a mismatch in rates).
If it is not available, or a proportion of lenders (35-50%) are unable to fund at the chosen benchmark rate, then Legacy Fall-back Options provide that benchmark rates should be determined in the following order:
Interpolated and extrapolated rates. In essence, statistically calculating new rates by plotting known rates and determining (i.e., estimating) what the rate would be by reference to the data that is available.
This may be achieved by several methods including:
- Shortening interest periods, and/or
- Using historic screen rates, or
- Using estimated (i.e. extrapolated) historic rates.
Reference bank rates. These are rates provided by a selected panel of “reference banks”.
Self-certified cost of funds. This is where a lender identifies a rate based on its cost of obtaining the funds to provide the loan. This may trigger a negotiation period with the borrower to agree an alternative basis.
The same as Option 1, but applies a simpler set of sub-options for the interpolated rates.
The LMA documents also currently include wording for determining rates for non-LIBOR currencies, a mechanism previously drafted to address the discontinuation of LIBOR rates for certain currencies in 2013. It requires a “Benchmark Rate” to be identified by the parties and adding schedules for specific market conventions (day counts, timing etc.). Although it is not a fully-formed solution it may be a useful device for using non-LIBOR RFRs (provided they align with a forward-looking term product).
Issues with Legacy Fall-back Options
The market universally recognises that the Legacy Fall-back Options are not a long-term solution. In particular, they do not address issues where a rate or a funding basis uses a different methodology entirely. That is, the initial fall-backs themselves are based on IBOR rates so they do not work once the IBOR is discontinued.
The Legacy Fall-back Options were designed for short-term disruption and/or the disappearance of an IBOR where any market replacement or substitute would be broadly analogous to an IBOR rate. The trigger points for the Legacy Fall-back Options provisions are typically where the relevant benchmark/IBOR is unavailable either temporarily (market disruption) or as a result of a permanent cessation. They do not allow for proactive change or for circumstances where a form of LIBOR or EURIBOR lives on, perhaps with a different methodology or calculation mechanism. So the parties may not be able to force a switch to a new benchmark where the relevant IBOR continues to exist but is no longer appropriate, or credible.
Polled rates (that is, reference bank rates) are no longer popular. Banks often do not want to be a reference bank, they can be administratively burdensome and, in any event, they may not work in a world where the relevant IBOR does not exist. There is no contractual obligation on reference banks to quote, and there will be inevitable disparity, so polled rates do not provide certainty.
“Cost of funds” is logistically difficult to determine and administer, particularly for large syndicates and the wider loan market, and for anything longer than the short-term. Furthermore, such an arrangement exposes the borrower to the credit standing of the lender which may have high funding costs (relative to a competitor lender).
The Legacy Fall-back Options drafting provides a template option for “Majority Lenders” and the Parent/Borrower to amend the documents to allow another benchmark rate to apply if a screen rate is not available (the LMA templates are a starting point only and the consent levels may vary from deal to deal). Lenders may have negative consent rights on revised benchmarks and may be deemed to have agreed to a new rate if they do not object within a limited timeframe. Lenders may also have additional consultation rights.
The amendment mechanism in the Legacy Fall-back Options is a relatively unsophisticated one and does not allow for a complete transition to a new rate or, more importantly, a change in margin (or for changes before LIBOR disappears from a screen), which are usually “all Lender” decisions. The change in margin mechanism is important in this context because even if a new rate (i.e., RFR) is agreed in respect of a facility, to be in an economically equivalent position to that before the introduction of the new rate, it is likely that the margin will need to be amended to account for the spread between the RFR and the pre-existing IBOR rate. This assumes that the new rate is not in itself a rate which is economically equivalent to the IBOR and near-risk free, for example, as with LIBOR+ or Zombie LIBOR.
Legacy Fall-back Options may not align with fall-back provisions in other documents or instruments. This may create mismatches in operations and payment obligations across interest rate swaps and other hedging arrangements. For the reasons set out above, the LMA has published revised optional wording for dealing with changes to the screen rate (see New transactions and the LMA Screen Rate Rider).
New transactions and the LMA Screen Rate Rider
The LMA’s recommended revised form of replacement screen rate clause (“Screen Rate Rider“) was published 25 May 2018 (the latest iteration, without substantive amendment was published 18 December 2018). It is designed for use with new transactions and is intended to introduce greater flexibility by enhancing the mechanism (currently included in its facility agreement templates) by which the parties to loan documents can agree a replacement benchmark. It does not seek to stipulate any replacement benchmark at this stage.
Rather it provides for:
- A wider range of amendments (including general alignment and consequential changes, implementing market conventions and adjusting pricing to reduce the gap between LIBOR and its successor);
- Expanded transition/discontinuance triggers (defined as a “Screen Rate Replacement Event”); and
- Revised consent processes for amendments, moving from an “all lender” position to a lower consent level being majority lenders plus obligors (and disenfranchising lenders who fail to respond).
The Screen Rate Rider is a stand-alone format, for use as slot-in clauses to allow parties to include it to the extent appropriate to their transaction.
It was agreed with ACT and other stakeholders, so theoretically reflects some borrower concern.
The Screen Rate Rider is expressly cognisant of the ISDA Benchmarks Supplement (see Derivatives) but does not directly echo the provisions due to the needs of different documentation and markets.
It is understood that although the option exists, lenders are not presently specifying SONIA or other RFRs as “Replacement Benchmarks” at this stage, due to the absence of term rates (although progress is being made on such rates, see RFRs).
There is some debate currently regarding the need for alignment of the consent threshold for determining a “Screen Rate Replacement Event” and that which applies to agreeing amendments to effect a replacement.
Clearly not all loans are based on LMA terms. Also many LMA-based loans will have been documented on older templates or modified wording, for example, historically some loans were documented on base rates (e.g. the Bank of England base rate) with associated floating-to-fixed swaps. In such loans there is no universal approach to using, or agreeing to use, an alternative rate should an IBOR be unavailable long term. It is more likely that bilateral loans would simply revert to a reference bank rate or cost of funds, which, as explained, may be problematic.
One example is the prevalence of IBOR based provisions in intercompany debt. Large organisations may have documented intercompany loans on an IBOR basis and such debt could run into the billions. While the solutions may be easier to resolve between the parties in an intercompany scenario, there is scope for IBOR issues on significant loan obligations to be overlooked when viewing the market through the prism of LMA terms.
Other documentation issues
The LMA and other market participants (including legal counsel), advise against trying to draft solutions beyond those provided in the Screen Rate Rider. This warning is generally based on two unwanted scenarios:
- That proposed drafting, in the absence of any clear RFR replacement for relevant IBORs, will make things more confused (where the methodologies/characteristics of those RFRs are unknown), and
- That for the most part, any solutions will be based on loose “agreements to agree” that will have little to no practical effect and/or limited enforceability as a matter of English contract law.
However there are many documents (both existing and future deals) that are either not written on LMA terms or will include fall-back provisions that do not align with the solutions offered by the LMA. In those scenarios and others, bespoke provisions will have to be crafted for both amending legacy documents and entering into new deals. In broad terms, the success of alternative fall-back language depends on whether they allow:
- The selection and application of alternative rates
- The calculation of any spread adjustments
- Consequential changes, at the relevant time
In the context of commercial agreements, there may be circumstances where drafting solutions that require transition to new RFRs or similar may simply be inappropriate. That is, LIBOR may historically have been used as a convenient stable benchmark without any further thought as to whether an interbank rate was necessary. It may be simpler in such transactions to use a base rate, for example. In all cases, the economic consequences may be significantly different to the existing position once the new rate is applied.
The LMA has indicated that it plans to amend its template documentation in due course to move away from IBOR benchmarks but that this will not happen until there is more certainty on how any replacement rates will operate and on more appropriate fall-backs.
Amending legacy documents preemptively, particularly for multi-currency debt, needs to be balanced with the practical reality of the absence of developed convention, methodologies and infrastructure for alternatives to the underlying IBOR benchmarks.
On the buy-side, purchasers of debt maturing after 2021 will need to look closely at the underlying loan documentation and the existence of any fall-back language. On the sell-side it may be increasingly difficult to trade debt that does not adequately address the need to transition to new benchmark rates after that date. Risk disclosures may need to be made. Marketing information, including information memorandums, may need to address how loans referencing LIBOR may operate through any discontinuance or transition period, or at least include appropriate disclaimers.
Secondary market transactions documented on LMA terms currently include their own definitions of LIBOR. In December 2017, the LMA’s Standard Terms and Conditions for Secondary Debt Trading were amended to add an additional fall-back provision to the “Relevant Benchmark Rates” to take into account IBOR discontinuance. Clearly the underlying benchmark rate is fundamental to the calculation of trades, including delayed settlement compensation, and it will be important to ensure that there is a consistent approach to alternative benchmarks spanning the primary and secondary markets.
US Loan Markets (LSTA)
The Loan Syndications & Trading Association (“LSTA“) is also a member of The Alternative Reference Rates Committee (“ARRC“) In keeping with the ARRC proposal, the LSTA has increasingly suggested that SOFR may also likely become the replacement rate for syndicated loans (and CLOs).
The current LSTA form provides for a fallback rate in the event LIBOR is not available – to an interpolated rate, an offered quotation rate to first class banks for deposits in the London interbank market by the Agent or a designated bank or a rate determined on the basis of quotes from designated “reference banks”. Credit agreements also typically provide that during a ‘market disruption event’, lenders may suspend their obligations to make LIBOR loans and that all outstanding LIBOR loans will switch over to a base rate or prime rate.
Because of this built-in fallback, syndicated loans (and CLOs) may be better situated to weather a transition than other products. However, since the prime rate is 200 bps above LIBOR, it may present a hardship for borrowers, so a long-term solution is still needed.
While the LSTA recommends paying particular attention to the fallback and alternative rates provisions in credit agreements it, like the LMA, recognises that, until widespread industry agreement has been achieved, it will be difficult to hard-wire a replacement rate into documentation.
A key consideration is that due to the innate differences between SOFR and LIBOR, in the event SOFR does become the prevailing replacement rate, it may behave differently than LIBOR has. As such, as has been highlighted in the ISDA transition report and consultation in the context of derivatives, credit agreements may also need to provide for a credit spread adjustment and term adaption.
As the market awaits consensus on fallback/alternative rates, ARRC and LSTA espouse a set of 4 guiding principles in dealing with the transition to a new reference rate for syndicated loans and CLOs:
- Fallback language should be incorporated into credit documentation as soon as possible, with an emphasis on flexible terms since a replacement reference rate has not been conclusively determined
- Consistency in fallback mechanism within and between asset classes should be emphasised – for example, it would be ideal for the fallback mechanism between a loan and a hedging product that protects the loan to function in the same way
- Fallback mechanisms should be practical and operationally feasible. They should also minimise value transfer and arbitrage opportunities
- The replacement reference rate should account for the importance of the rate, spread and term structure.
While the LSTA has not officially recommended a replacement reference rate or proposed new fallback language, it has repeatedly stressed the need for immediate attention to these provisions in current credit documentation. It acknowledges that language currently being negotiated must remain flexible since there is no industry consensus as to the replacement rate, but that there is still value in wiring in a streamlined amendment process. As replacement mechanisms crystallise, the language will likely evolve to be more specific and granular.
Frequently Asked Questions
Q. What should we be doing about our existing contracts and arrangements?
Due diligence should identify matters such as:
- (a) what calculation mechanics are used?
- (b) what rate fall-back provisions (if any) are used, including identifying and analysing trigger events and who selects the replacement rate?
- (c) what amendment provisions are there, including consent thresholds, instructing groups and authorisations including in intercreditor structures and the scope for amendment flexibility?
- (d) what interdependencies are there with other products and instruments?
- (e) can loans be repaid or prepaid if no agreement can be reached on IBOR replacement?
Once documents are identified and analysed, it may be necessary to put in place processes to effect appropriate amendments to those arrangements (including aligning arrangements where possible to avoid basis risk). In some instances the arrangements may be refinanced or revisited within the timeframe for transition. However other documents may need to be amended more proactively.
In each case any amendments should pay close attention to, among other things, the progress of development of any replacement benchmark rates, the prevailing market view and the respective commercial interests of the parties. It is almost inevitable that there will be significant time and cost implications in any repapering exercise.
Q. What should we be doing about new transactions?
Before drafting provisions that contemplate new rates with different calculation methodologies and conventions, an analysis should be done as to whether the parties are set up operationally to work with any replacement benchmark, before committing to its use, even in fall-back scenario.
Where new loans or agreements will include payment obligations that extend beyond the likely discontinuation date in 2021, the parties should consider to what extent they need to future-proof any change in interest rate benchmarks.
For LMA-based documents this will likely come down to a consideration of the extent to which the Screen Rate Rider, including the various options within those riders (see New transactions and the LMA Screen Rate Rider), should be incorporated in the loan agreement.
For non-LMA documents a more bespoke solution may be required that more closely reflects the respective commercial interests of the parties.
However some market commentary has advocated an approach that avoids continuing to use IBOR benchmarks, for the time being, with enhanced fall-back provisions, preferring to draft agreements without any reference to IBORs. However this remains problematic in instances where a suitable alternative rate is not yet available. In most instances a future “amendment” option will be required and this is likely to have time and cost implications.
Q. Why can't we just say "if LIBOR is discontinued then the rate shall be SONIA"?
Q. Why do we need a forward-looking term rate?
Q. What about the potential for mismatches in e.g., hedging arrangements and other financial instruments?
Q. Will lenders and borrowers have to change their processes, operations and systems?
Check list of things to consider
- Develop plans to mitigate the risks of IBOR discontinuance and reduce the dependencies on IBORs (across all products, not just loans).
- Review existing documents (legacy contracts) and the way your business uses IBORs. Consider the full range of products/instruments and the need to align your approach across swaps, bonds and loans.
- Consider approaches to documenting new transactions with maturity dates or payment obligations beyond 2021, including where legacy deals are amended/extended to mature post-2021.
- Consider what consent thresholds are appropriate for documenting any changes to benchmark rates (both in amendments to legacy contracts and also documenting new transactions).
- Actively monitor developments and industry initiatives.
- Consider participating in industry working groups and responding to consultations.