October 18, 2018 - On September 24th, the Alternative Reference Rates Committee (the ARRC) released a consultation on LIBOR fallback language for syndicated loans. On October 4th, the LSTA hosted a webcast that detailed the consultation. We also are releasing a series of briefs that explain the fallback consultation to help market participants respond to it. (The LSTA was one of the drafters, so we are very familiar with it.)
Last week, the LSTA focused on the triggers that initiate the transition from LIBOR to a new reference rate. This week, we turn to the proposals themselves. The ARRC LIBOR fallback consultation offered two proposals for fallback methodology, the “Amendment” approach (which is similar to what is in the market today) and the “Hardwired” approach (which we deconstruct below and which is discussed beginning at minute 21 in the LSTA Webcast).
The foundational principle of the Hardwired approach is that all the terms of LIBOR transition are “hardwired” into the credit agreement upon origination. Because everything is defined upfront, ideally there is no need for an amendment at transition – which may be challenging if thousands of loans must be transitioned simultaneously – and it reduces potential for gamesmanship at transition.
Here is how the Hardwired approach works: At the initiation of the credit agreement, the lenders and borrower agree to i) the triggers, ii) a replacement rate waterfall and iii) a spread adjustment waterfall.
As discussed last week, there are five mandatory triggers that force a conversion from LIBOR to the new reference rate. They are triggers that indicate LIBOR cessation, an unannounced stop to LIBOR, a decline in the number of banks that make LIBOR submissions or the relevant regulator saying that LIBOR may no longer be used.
Once a trigger occurs, the Hardwired approach immediately turns to finding a replacement rate by using a replacement rate waterfall (see the Hardwired’s “Replacement Benchmark” definition in the draft contract language on p. 25 of the consultation). The Hardwired approach first falls back to forward-looking term SOFR. Because forward-looking term SOFR looks like LIBOR, e.g., it should have one- and three-month tenor options, it is viewed as the “best” fallback. However, forward-looking term SOFR does not currently exists and if it still does not exist when the trigger occurs, we move to the next rate in the waterfall, Compounded SOFR. Compounded SOFR still provides a term rate for the loan and, if the rate is compounded “in advance”, it will be known for the life of the loan. (In contrast, a rate that is compounded “in arrears” would not be known until shortly before the interest rate is due.)
If for some reason, Compounded SOFR does not exist when the trigger occurs, we move to the final stage in the rate waterfall: Overnight SOFR. This rate exists today.
But we are not done yet. SOFR is a secured, risk-free rate, whereas LIBOR contains an element of bank credit risk. For this reason, term SOFR is expected to be lower than term LIBOR. Thus, we look to add a “Replacement Benchmark Spread” (p. 26 of the consultation) to SOFR to make it more comparable to LIBOR. The first stage in the spread waterfall is an adjustment (or a method of calculating an adjustment) that has been selected, endorsed or recommended by the Federal Reserve or the ARRC. If that does not exist when the trigger occurs, then we move down the waterfall to a spread or calculation methodology that has been selected by ISDA as its fallback for derivatives. (ISDA currently is out with a consultation on its spread methodology.) Do note that there was discussion about whether to include a fixed spread fallback that was, effectively, equivalent to “LIBOR minus SOFR”. However, the ARRC ultimately decided not to include the option as it could be volatile or subject to manipulation if LIBOR were deteriorating.
Having explained the Hardwired approach, the consultation next asks a series of questions about it. After showing a chart (p. 6) illustrating the triggers, fallbacks and amendments in both the Hardwired and Amendment approaches, the consultation asks a foundational question, “Which approach would you use and, if you would prefer the amendment approach today, what specific information would you need to adopt a hardwired approach?” The ARRC simply wants to know which approach has more traction – and why. Moreover, as the Hardwired approach may be easier to implement en masse and thus reduce the potential for market disruption, the ARRC wants to know what additional information would be useful to facilitate adoption.
Next, because the Hardwired approach determines the replacement rate using a waterfall, Questions 5-11 deal with replacement rates. (We discuss this starting at minute 32 of the webcast.) First, on pps. 9-10, the consultation explains that the ARRC intends to endorse a forward-looking term SOFR rate, assuming that an IOSCO compliant version can be created. However, derivatives are expected to fall back to an overnight rate or a compounded overnight rate, potentially creating a mismatch with hedges. Question 5 addresses this, asking “Should a forward looking term rate be the primary fallback, even though derivatives are expected to reference an overnight version of SOFR?”
Next, while there should be a forward-looking term SOFR, it is possible that certain periods – like 6-month SOFR – might not exist. If that is the case, Question 6 of the consultation asks “Should the administrative agent be able to eliminate certain interest periods if there are no equivalent SOFR terms available?”
If forward-looking term SOFR does not exist, then the Hardwired approach moves down the rate waterfall to Compounded SOFR (pps. 10-11). There are two options for Compounded SOFR. The first is “Compounded in Advance”, which would mean that, upon origination of the loan, SOFR would be calculated looking back over one month or three months and thus the rate would be known in advance of the interest period. The second option is “Compounded in Arrears”. In this case, SOFR would be compounded while the loan was outstanding and therefore would not be known in advance. Question 7 asks, “Should Compounded SOFR be the second step in the waterfall?” and Question 8 asks, “Should Compounded SOFR be compounded in advance or in arrears?”
The third stage of the Hardwired waterfall is overnight SOFR. In this proposal, SOFR would be set on one day at the beginning of the interest period and would remain in effect for the duration of the interest period. Question 9 asks, “Is Overnight SOFR an appropriate fallback or should the final replacement reference rate be Compounded SOFR?” Question 10 asks, “Is fixing one observation of overnight SOFR for a term that could last three months or longer appropriate?”
Finally, there is a catchall Question (11) that asks whether another replacement rate should be added to the hardwired approach waterfall.
The next step in the Hardwired waterfall is a static spread adjustment to make LIBOR and SOFR more comparable. The Hardwired approach offers two potential spread adjustments. First, the Fed or the ARRC is considering endorsing a spread adjustment or methodology specifically for cash markets. Question 12 asks, “Should ARRC recommend a spread adjustment for cash products like loans?”
If ARRC does not offer a recommended adjustment, the next stage in the spread waterfall is an ISDA recommended spread adjustment. Importantly, inasmuch as ISDA is looking to fallback to an overnight SOFR, this is unlikely to be a perfect match if loans fallback to a forward looking term rate. Bearing the differences in mind, Question 13 asks, “Should an ISDA spread adjustment be the second stage of the spread adjustment waterfall?” Finally, a catchall Question 14 asks whether any other spread adjustment should be considered.
Importantly, if one goes down the waterfall and still cannot lock in a replacement rate plus spread adjustment, the Hardwired approach will fall back to an Amendment approach. Thus, there is always a failsafe mechanism.
While there are many details – and hence many questions – in the Hardwired approach, this is because it attempts to define all potential situations in advance. This approach removes discretion (and potentially gamesmanship) and potentially allows for the rapid transition of loans en masse. However, the questions in the consultation attempt to determine if the Hardwired approach is ultimately workable. We hope to hear your opinion!