Next Wednesday, September 30, 2020, marks the ARRC’s recommended transition date to a “hardwired” approach for LIBOR successor provisions in U.S. dollar-denominated syndicated credit facilities. It also represents the 15-month deadline before LIBOR is to be completely phased out. On the cusp of these pivotal horizons, we assess what you need to know from a fund finance perspective for the next chapter of the LIBOR transition.
But first, let’s start with where we are today. LIBOR transition language first started making its way into credit agreements as far back as fall 2017, and by late 2018 the New York Fed’s Alternative Reference Rates Committee (ARRC) began publishing consultations on LIBOR fallback language for business loans. In the time since, a majority of fund finance market participants have tended away from the hardwired track and instead adopted ARRC’s alternative: the “amendment” approach. This regimen provides trigger events for when a fallback is to be selected by the parties, a mechanism for negotiating the new benchmark rate and provisions to make the replacement rate more comparable to LIBOR. Yet it permits all decisions about the actual successor rate and spread adjustment to be made in the future when LIBOR becomes unavailable.
Either option is voluntary. While the hardwired course is intended to offer certainty for what the future benchmark rate will be, several factors have made the amendment approach the more popular route in our market until now. Initially, it was the only option. The syndicated business loan market has assessed potential replacements for LIBOR since 2014, but without a generally accepted successor until relatively recently, market participants have chosen the more flexible amendment scheme to delay selecting a definite alternative. With LIBOR not expected to fully phase out until the start of 2022 and stated tenors in subscription credit facilities most often ranging from just one to three years, parties in our market did not need to implement a more certain solution sooner. Plus, the amendment language that has been adopted almost ubiquitously includes safeguards requiring the administrative agent and borrowers to duly consider prevalent market practices when the time does come for selecting a replacement rate.
That time is soon approaching. As we get closer to the end of LIBOR, market participants will have a greater understanding of how to operationalize a successor benchmark. So they should adopt more specific language for selecting fallback rates and spread adjustments that gives greater certainty to their counterparties and facilitates a consistent transition across their portfolios. Thus in June, ARRC refreshed its previously suggested hardwired provisions to do just that. Below is a brief primer on ARRC’s recommended benchmark replacement, SOFR, followed by a detailed description of the updated hardwired approach.
SOFR
SOFR is the secured overnight financing rate for transactions in the Treasury repo market. SOFR is administered by the New York Fed and has been published on its website since April 2018. It was selected from among various possible fallbacks to LIBOR principally for the robustness of its observable transactions. Because current outstanding contracts that reference U.S. dollar LIBOR have been estimated to aggregate almost $200 trillion in value, ARRC sought a successor benchmark that would provide representative value. On a daily basis, SOFR indexes to a goliath volume of roughly $1 trillion worth of U.S. dollar transactions in the repo market. LIBOR, on the other hand, is based on a relatively miniscule amount of about $500 million in daily London interbank loans.
The largest difference functionally between the two for subscription credit facilities is that SOFR is an overnight rate, so it updates daily. LIBOR generally is a term rate that may be quoted in one-month, three-month, six-month or other tenors. SOFR is also a secured rate, while LIBOR is unsecured, so they behave differently in different scenarios. To determine LIBOR, one only needs to obtain the applicable rate from a vendor screen page two business days before the start of the interest period, and the rate applies to the entire term. Conversely, SOFR is quite complex, and the methodologies to calculate and implement it are still developing for use in different markets. As further described below, there are at least four variations of SOFR that are in development. Our market is looking to evolve SOFR to make it work similarly to LIBOR, both operationally and in our legal documentation. Accordingly, ARRC’s new hardwired language seeks to bridge the gap between these two benchmarks.
Hardwired Approach
The basic template for the hardwired approach is that the lenders and borrowers agree to the following:
- The triggers that will lead to LIBOR being replaced by a successor rate;
- A waterfall-like series of steps by which that successor is selected;
- A similar waterfall for the spread adjustment that will make the lenders’ margin remain consistent so the replacement benchmark is more similar to LIBOR; and
- The ability to make conforming changes so that the end result functions as intended.
Triggers
The first step is similar to the first stage of the amendment route. A new successor benchmark will be selected on the occurrence of one of four triggers, as set forth in ARRC’s definitions of “Benchmark Transition Event” and “Early Opt-In Election.” Those triggers can be the administrator of LIBOR publicly announcing that LIBOR has ceased or will cease; a similar public notice by an applicable regulatory supervisor, central bank or resolution authority; a regulatory supervisor publicly disseminating a determination that LIBOR is no longer representative of the underlying market; or an early opt-in election by the parties to the credit agreement.
Successor Waterfall
After a trigger occurs, the amendment approach diverges by leading to a process for negotiating the new benchmark. The hardwired provisions, on the other hand, move to find a replacement rate based on a successor waterfall pursuant to ARRC’s defined term “Benchmark Replacement.” The first such successor that would be selected is Term SOFR. Term SOFR is conceptually the simplest version of SOFR because it functions most similarly to LIBOR. It provides a forward-looking rate that is to be applied at the start of an interest period and used for that entire term, whether it be a tenor of 30 days, 90 days, 180 days or otherwise. There is only one significant issue with Term SOFR today: it has not been created yet. There is also no certainty that its conception will be advanced enough by the end of 2021 to adequately support the market.
If Term SOFR is unavailable when LIBOR is being phased out, the replacement waterfall next flows to Daily Simple SOFR as the chosen rate. This replacement benchmark is structured to work much like Daily LIBOR and is akin to how a reference rate, prime rate or federal funds rate would be determined: by multiplying the SOFR rate published on the New York Fed’s website by the principal amount of outstanding loans for the applicable day. Daily Simple SOFR is intended to be easy for lenders to administer. It also has a fairly low basis to the SOFR formulation endorsed by the International Swaps and Derivatives Association (ISDA), so it can be hedged by SOFR-based derivatives.
While Daily Simple SOFR is ARRC’s recommended secondary fallback if Term SOFR is nonexistent, each institution implementing a LIBOR successor will want to analyze its own capabilities and propensities when determining these replacements. Some firms may prefer to adopt Compounded SOFR in Advance. Of the current SOFR variants, this one operates most like traditional LIBOR. A rate is simply pulled from the New York Fed’s web page and applied to the next interest term, which makes a transition to this benchmark perhaps the easiest to execute. The primary concern with Compounded SOFR in Advance for the general loan market is that it is backward-looking, as it is derived from SOFR averages for the prior corresponding tenor, and so it may not accurately reflect the rate for the next interest period of the same tenor.
Other market participants might choose to use Compounded SOFR in Arrears in their fallback waterfalls. With this benchmark, the interest rate is computed at the end of an interest period by determining what SOFR was on each day of the period and then compounding it for that term. Accordingly, Compounded SOFR in Arrears is the most economically accurate of the SOFR types. One challenge with Compounded SOFR in Arrears is that it is quite complex to calculate and manage operationally. The current methodology would also need to be further developed before it can effectively accommodate loans that may have principal change over time.
The last step in ARRC’s benchmark waterfall kicks in if a replacement rate cannot be determined using one of the SOFR alternatives specified in the first two stages of the waterfall. In that case, the parties shift to a streamlined amendment process that mirrors the current amendment approach. The administrative agent and the borrowers are to select a rate with due consideration of any market conventions or recommendations by ARRC for determining a replacement at that time. If the required lenders object to the proposed alternate rate, then the amendment fails and the process starts again, with the interest rate being tied to the reference rate or base rate in the meantime.
Spread Waterfall
To make SOFR’s rate level be more comparable to that of LIBOR, the next part of the hardwired approach deals with a spread adjustment. The adjustment is intended to be tenor-specific. So a spread adjustment for one-month LIBOR would be applied to one-month SOFR, as would going from three-month LIBOR to three-month SOFR, and so on. Such adjustment is selected through a waterfall under ARRC’s definition for “Benchmark Replacement Adjustment” that is much like the waterfall for the replacement rate itself. In each case, the adjustment could be a positive number, a negative number or zero.
The first two steps of this waterfall only apply if the benchmark replacement that has been chosen is a specified SOFR fallback, such as Term SOFR or Daily Simple SOFR. First, the waterfall provides for a replacement adjustment that is recommended by ARRC. ARRC’s process for selecting a spread adjustment is based on the median calculated difference between SOFR and U.S. dollar LIBOR over the prior five-year period. This is consistent with ISDA’s method for determining the fallback adjustment for derivatives. However, if ARRC has not selected a replacement adjustment when LIBOR is to cease, then the waterfall’s second step would lead directly to using ISDA’s adjustment. Both ARRC and ISDA have announced that they expect to publish their recommended spread adjustments before any LIBOR pre-cessation or cessation event occurs.
If the replacement rate is instead selected by the amendment process at the end of the successor waterfall, then the last step of the spread waterfall applies. As in that amendment stage, this third step requires the administrative agent and the borrowers to choose an adjustment while giving due consideration to market practice or ARRC’s recommendations for selecting a fallback adjustment.
Conforming Changes
Although the hardwired provisions are intended to give a definite rate and adjustment for the LIBOR successor, it is likely that certain general updates will need to be made operationally or in the loan documentation to ensure an effective transition to the new benchmark. ARRC’s definition of “Benchmark Replacement Conforming Changes” provides for such modifications. It gives a non-exhaustive list of the types of technical or administrative changes that may need to be made by the administrative agent, whether during the transition or as needed afterward.
Multicurrency Facilities
One asterisk on the hardwired approach is that it is limited to U.S. dollar-denominated loan facilities. Each other jurisdiction with a LIBOR-quoted currency is determining its own recommended replacement rate. With LIBOR, the foundation for the interest rate was the same across each alternative currency; they were just priced in distinct denominations. But because these currencies will be based on different rates after LIBOR ends, each of their subsequent benchmarks will require its own margin to maintain consistency for comparison to LIBOR. This will add complexity to the documentation and administration of multicurrency credit facilities.
Final Thoughts
Our market wants efficiency for the transition away from LIBOR while also ensuring equitable outcomes. Lenders and borrowers both want to ensure their margins remain consistent with what they have been under LIBOR. There is still no market consensus for what the replacement benchmark will ultimately be. Each institution must determine what works best for it. As with ARRC’s previously suggested provisions, the current formulation of the hardwired approach is meant to be the next phase in an iterative process as our market approaches the LIBOR sunset.
When will all of this be implemented? ARRC’s recommended best practices for LIBOR transition includes the following suggested schedule for business loans like those found in our fund finance market:
- September 30, 2020: Hardwired fallbacks are incorporated for syndicated transactions.
- October 31, 2020: Hardwired fallbacks are included for bilateral deals.
- June 30, 2021: Target date for the end of new originations using U.S. dollar LIBOR.
For more information, here are links to (1) ARRC’s updated hardwired approach and user guide, as published on June 30, 2020, and (2) ARRC’s recommended best practices with implementation schedule.