You are a corporate treasurer of a company with debt maturing now or over the next few years, and the company may need to increase its leverage during that period to finance new projects. You are considering the company’s options in the debt markets and recall hearing this incessant drone about LIBOR. What is really going on, and how should it affect your thinking?
Unfortunately, trying to make sense of LIBOR can be like trying to take a drink from a fire hose. LIBOR is deeply embedded in the financial world, and many regulatory and industry groups are working on a global transition from LIBOR to alternative interest rates in many currencies, markets and financial products. Further complexity lies in considering aspects of the issue beyond what debt contracts say, including regulatory, tax and accounting aspects. There is a great deal of work product for borrowers to sort through and, despite coordination efforts by the various groups, much of it is not currently in alignment.
To briefly summarize, LIBOR really is expected to go away at the end of 2021, and financial regulators globally have urged all companies to recognize this and factor it into their planning. The chief reason for this is that the short-term interbank funding markets, which historically served as the underpinning for the bank rate quotations from which LIBOR was calculated, have shrunk dramatically. Many LIBOR quotes have been based on estimates, or “expert judgment,” due to the lack of actual trades on which to base quotes. Regulators believe that the small size of these markets made LIBOR more susceptible to the manipulations that came to light during the financial crisis. Further, there is an estimated US $350 trillion of outstanding financial contracts that refer to LIBOR, and regulators have expressed concern with having such a large volume of contracts refer to prices derived from a thinly-traded, much smaller market.
Regulators believe that the inherent weakness of LIBOR may lead market participants to conclude that it is unreliable at some point in the future (recall that LIBOR’s perceived unreliability caused significant problems when the financial crisis was at its most severe). Regulators are further concerned that banks may decide to stop submitting quotes due to either the lack of underlying transactions or to avoid liability. Any cessation or perceived unreliability of LIBOR could cause massive disruption.
In 2017, the UK Financial Conduct Authority (FCA) announced that it would not use its power to persuade or compel banks to submit LIBOR quotes after the end of 2021, but also announced that it had secured commitments from the LIBOR panel banks to continue submitting quotes through then. That announcement marked the beginning of a large, global project to come up with interest rate replacements for LIBOR which are better than LIBOR, to bolster contractual language and to develop solutions for the many second-order issues arising from these first-order changes. The deadline remains the same and, though considerable progress has been made, the project is not complete and time is growing shorter. Further, despite all of the regulatory activity to date, many market participants are frustrated at the amount of uncertainty and lack of clarity that remains.
So what does all this mean for you?
On the positive side, regulatory bodies have identified “risk-free” rates (RFRs) as interest rate alternatives in all the LIBOR currencies (US dollars, sterling, euro, yen and Swiss francs). While the basis of calculation of these rates differs, these rates are all similar in that they are overnight rates, and do not reflect term risk or counterparty credit risk, thus making them very different from LIBOR, which is a term rate and factors in both of these risks.
These RFRs were developed to meet new regulatory standards for interest rate benchmarks put into effect since the financial crisis, are based on large volumes of actual transactions, and are designed to reduce the opportunity for conflict of interest and manipulative conduct that plagued LIBOR. These regulatory standards were developed by the International Organization of Securities Commissions (IOSCO), have been endorsed by the global Financial Stability Board (FSB) and are reflected in the EU Benchmarks Regulation (BMR). To be clear, LIBOR (although now reformed) has been judged to not meet these standards.
These RFRs are at different stages of development for the various LIBOR currencies. The RFR for US dollars, SOFR, began to be quoted in 2018. The RFR for sterling, SONIA, has existed since 1997, and perhaps as a result there has been considerably more debt volume denominated in SONIA and aggregate notional amount of cleared SONIA swaps than corresponding amounts in SOFR. Trading in the RFRs for euro, yen and CHF, €STR, TONAR and SARON, lags further behind. To complicate matters, in Japan and the euro zone, the regulators have opted for a “multiple rate” approach for interest rate benchmarks. This sees their reformed and improved local IBORs, TIBOR and EURIBOR, set to co-exist with the identified RFRs for their currencies. €STR was launched on 2 October 2019, and replaces EONIA (with EONIA now to be quoted at €STR plus a fixed 0.085% spread from that date until being discontinued completely on 3 January 2022).
The track records for those RFRs that have been available indicate that they may behave differently than LIBOR, particularly during periods of stress.
Many swaps traders have said that they are pleased with these RFRs, and place less of a priority on the development of other interest rates.
Corporate borrowers, on the other hand, have expressed a desire for forward term rates that would replicate LIBOR’s current maturity structure. While there has been an appreciable volume of debt instruments issued that is denominated in RFRs, notably floating rate notes (FRNs), these instruments calculate interest in arrears, on a compounded or simple average basis (often using a “lag” or “lockout” mechanism). Many corporates like the certainty provided by LIBOR for setting rates and planning cash flows in advance.
Forward term rates based on the RFRs would address this. However, these forward term rates will also need to comply with the new benchmark regulations, and be based in a sufficient volume of actual transactions to justify forward term benchmarks. The amount of liquidity in the RFRs, and in transactions supporting robust forward yield curves based off RFRs, is therefore critically important to developing these forward term benchmarks.
It is unclear (and perhaps tending to doubtful in some cases) whether these forward term benchmarks will be developed before the end of 2021, although such rates for some currencies may be closer than others. The Swiss regulators announced at the end of 2018 that it was not currently feasible to develop such a rate for CHF. The global regulators have urged market participants to press on with LIBOR transition without waiting for these rates to arrive.
As noted above, there has been notable evidence of a switchover in the FRN markets to RFRs, particularly in SONIA, the sterling RFR. The growing popularity of these RFRs with FRN issuers and investors has been buoyed by the publication of literature on how to use the RFRs in new FRNs.1
However, companies continue to issue debt that refers to LIBOR and matures after 2021. In fact, we are not aware of any syndicated credit facilities in EMEA, the US or in Asia Pacific that refer to RFRs.
For LIBOR-denominated debt (both new issuances and legacy debt), contracts have focused on the insertion of fallback provisions to apply if LIBOR goes away. Two basic approaches have been put forth: (i) an “amendment approach” that provides a mechanism for borrowers and lenders to negotiate and implement a replacement rate by means of an amendment to a debt contract in the future, and (ii) a “hardwired approach” that implements a replacement rate without the need for a future amendment to the debt contract based on triggers, terms and conditions agreed to upfront.
Under the amendment approach, if a trigger event occurs and for some reason an amendment is not agreed to, in most cases LIBOR ceases to be available as a pricing option, and the loan would be priced at the cost of funds rate (or, in the US, at the base rate).
In the US, the Alternative Reference Rates Committee (ARRC) has recommended fallback provisions for US dollardenominated loans (both syndicated and bilateral), FRNs and securitizations. The ARRC’s loan fallbacks offer a choice between the amendment approach and the hardwired approach for loans, but only the hardwired approach for FRNs and securitizations; the amendment option for loans reflects the fact that loans are relatively easier to amend than the other debt products. The ARRC’s hardwired approaches all use waterfall provisions to identify a replacement interest rate; we note that the top priority in each waterfall is a forward term SOFR that has been recommended by the ARRC, which rate does not yet exist.
The ARRC hardwired approaches also use waterfall provisions to identify an appropriate credit spread adjustment between LIBOR and the replacement interest rate, to account for the lack of counterparty credit spread in the replacement rate; we note the top priority in each waterfall is a spread adjustment that has been recommended by the ARRC, which spread adjustment does not yet exist.
The LMA in EMEA and the APLMA in Asia Pacific have each recommended a variation of the amendment approach in their forms of facility documentation for loans. The Association of Financial Markets in Europe (AFME) has proposed hardwired fallback wording for inclusion in European securitization documentation.
To date, most LIBOR bank facility agreements we have seen have used some variation of an amendment approach, frequently requiring borrower consent and the consent of majority (not all) lenders. We have seen some evidence of adoption of the ARRC hardwired approach for fallbacks in LIBOR FRNs and securitizations.
It is likely too early to tell whether any of the proposed hardwired fallbacks will be broadly accepted.
While this article focuses mainly on debt products, we note that the International Swaps Dealers Association (ISDA) has done yeoman’s work in considering many of these issues as they affect swaps, including developing their own fallback provisions and working on methodologies for credit spread and term adjustments.
The ISDA work has greatly informed the work of working groups for other products and markets. However, while ISDA and the working groups have tried to coordinate their efforts, they serve different masters, and the masters they serve have differing interests. Although a “one size fits all” solution for LIBOR would certainly be appealing, it is far more likely that market participants will need to carefully evaluate the effect of various bespoke fallback provisions and other developments in transactions involving several asset classes, such as CLOs and other ABS transactions, hedged loans or standalone rate swaps, to ensure that they understand any basis risk arising from the fallback and other provisions not being aligned.
So where does all this leave the corporate treasurer seeking to obtain and price debt financing now?
For bank deals, this likely continues to mean pricing now at LIBOR, with an amendment approach fallback. Depending on the currency involved, it may be possible for a company to issue RFR-denominated FRNs now, with interest calculated in arrears, consistent with newly developed market conventions. For LIBOR debt with a fallback that remains outstanding, an event will very probably occur with respect to LIBOR (at the end of 2021 or possibly before, if LIBOR is judged unrepresentative of underlying financial reality by the FCA, LIBOR’s principal regulator) that triggers the fallback mechanism, and at that time it is hoped there will more clarity as to what the market has come up with as a replacement rate.
However, if a fallback trigger were to occur today, the rate supplied by the ARRC hardwired approach would be either (depending on the selection made when the fallback was put in place) compounded SOFR or a simple average of SOFR, calculated in each case in arrears, with a spread adjustment. The rate that would be arrived at under the amendment approach, if a trigger occurred today, is less certain and of course subject to the agreement of the parties in each case (assuming the borrower has a consent right). However, there is good reason to think that a compounded or average rate derived from an RFR, calculated in arrears, would loom as a likely option, since there is not yet an IOSCO-compliant forward term rate for any LIBOR currency. The borrower could seek to refinance if it disagreed with the rate the lenders were proposing, but might not find a better deal (and keep in mind that the markets might then be crowded with borrowers seeking a similar deal). Even if the borrower had a consent right, that might be cold comfort if LIBOR were to cease being available and the loan were to default to pricing at cost of funds or base rate until an amendment was agreed.
The development of a forward term benchmark based off an RFR prior to the occurrence of a fallback trigger would make that benchmark an option that many borrowers would prefer (assuming the economics were equivalent). In the ARRC hardwired fallback, if the ARRC were to recommend a compliant forward term benchmark for US dollars, that rate would take priority in the waterfall over compounded/simple average SOFR in arrears.
It may be possible for a company to hedge LIBOR transition risk through a derivatives strategy. We note that many financial institutions see significant arbitrage opportunities in this area.
It is possible that, between now and the end of 2021, loans will begin to be denominated in RFRs, and that market conventions will be developed for those loans (which likely will be derived from the conventions for RFR FRNs). We note that the LMA recently published exposure drafts of facilities agreements for SONIA and SOFR compounded in arrears2. In addition, the LTSA is working on a concept audit agreement referencing a compounded average of SOFR calculated in arrears.
It is also possible that forward term benchmarks derived from RFRs will be developed prior to the end of 2021 that will enable bank facilities, FRNs, securitizations and other products to be priced by reference to such benchmarks. Treasurers may not want to lay odds on this occurring, although it is a reasonable expectation that sufficient liquidity will build in sterling and US dollars that at some point will enable forward term benchmarks in those currencies that comply with the benchmark regulations. The development of forward term benchmarks in the other LIBOR currencies may take still more time.
Fallback provisions for loans may migrate from the amendment approach to the hardwired approach, particularly if more progress is made in developing actual rates and spread adjustments that would fill in the steps in the waterfall provisions. On a macro level, moving to a hardwired approach would probably be a good thing for the debt markets, since it would mitigate the risks of having to amend a large number of credit facilities at the same time and possibly coming up with inconsistent results.
The RFRs may not reflect bank funding costs, further distinguishing them from LIBOR (which is a cost-plus pricing model). This could lead to banks seeking compensation for these costs through increased spread.
Despite all this, there may be some chance (which may not be more than a forlorn hope) that LIBOR will survive, in some form, after 2021. In order to be viable thereafter, it would need to be revised substantially to comply with the benchmark regulations. Regulators have warned firms to not count on LIBOR’s survival and have pointed out that, if it does survive, it may not be compliant (and, if not compliant, banks subject to the BMR would not be able to use it), may not be supported by bank submissions, may not be reliable and may be so changed from its current form as to not be the same thing.
Work continues on a large number of separate workstreams in LIBOR transition. There will be further developments. We expect that progress will continue to be incremental (as well as uneven among currencies, markets and debt products), and that there likely will be no single defining moment when the scales will fall from the market’s eyes and all will be clear.
One of the guiding principles that the regulators and working groups hoped to adhere to in the transition process has been the avoidance of value transfer from one side of a trade to the other. Regrettably, that goal may prove elusive.
As always, treasurers should keep their refinancing options open and an eye out for market developments.
We note that the above discusses only a part (though an important one) of the enormous headache that LIBOR transition poses for corporate treasurers, and that, because LIBOR is so deeply embedded in financial contracts, treasurers will very likely need to do much more to understand and manage the total transition risk of their firms. But the discussion of the other parts will need to take place elsewhere.3
1 Examples of this include: the Sterling Working Group’s Conventions for referencing SONIA in new contracts (March 2019); the following publications from the US Alternative Reference Rates Committee A User’s Guide to SOFR and Recommendations regarding more robust fallback language for new issuances of LIBOR floating rate notes (both April 2019) and SOFR Floating Rate Notes Conventions Matrix (August 2019), and the FSB’s Overnight Risk-Free Rates – A User’s Guide (June 2019).
2 LMA subscribers may access the drafts here.
3 See, e.g., the ARRC’s Practical Implementation Checklist for SOFR Adoption. For more on LIBOR transition, see the Baker McKenzie LIBOR Hub page.