Term Versions of Libor-Replacement Rates Pick Up Steam

Recent input from market participants about their preferred way to transition to a Libor-benchmark replacement surfaced concerns about several interconnected risks. Necessary to address them will be the development of a forward-looking version of the new risk-free rates (RFRs), an effort that began in earnest in October and is anticipated to gain momentum in 2019.

In December, the Alternative Reference Rates Committee (ARRC), which chose the Secured Overnight Financing Rate (SOFR) to replace Libor for dollar-denominated transactions, posted on its website comments regarding consultations for developing fallback language for syndicated loans and floating-rate loans. It also launched a new consultation on fallback language for bilateral business loans and securitizations, with comments due Feb. 5.

The fallbacks are contractual language designed to transition products from Libor to an alternative benchmark. Such language is critical for existing transactions that mature later than 2021, after which banks will no longer have to provide interbank-lending-rate submissions, or should ICE Benchmark Administration (IBA) no longer have sufficient data to publish it.

The loan-syndication consultation's first question, and the key to resolving other important issues, is whether there should be a “hardwired” methodology that is essentially set in stone, transparent, and reliable, or if an amendment approach providing more flexibility is preferable.

The ARRC defined the latter as providing a “streamlined amendment mechanism for negotiating a replacement benchmark in the future.” It notes that the amendment approach could potentially serve as an initial step toward a hardwired approach, which would more clearly explain how to identify and implement a potential replacement rate.

Twelve of 20 signed comments on the syndicated loan consultation - mostly from lenders and a few large insurance companies - supported the hardwired approach. JPMorgan Chase & Co. describes several advantages, including the consistency and uniformity of SOFR across different credit facilities and products, the certainty of the replacement rate ahead of the actual replacement date, and the operational benefit of not having to negotiate and enter into potentially thousands of contractual amendments.

Prerequisites for Hardwired

For a hardwired approach addressing interbank offering rate (IBOR) fallbacks to work, certain key elements need to be in place soon: The alternative rates themselves have to be developed and gain liquidity; standard methods for calculating the spread adjustment between the old and new rates need to be developed; and both the new rates and spread adjustment methodologies need to gain widespread market acceptance. A spread adjustment between Libor and the RFR is necessary to avoid a value transfer when it comes time to switch rates.

One current challenge to hardwired approaches is that proposed fallback language for debt and derivatives can fall back to different rates. For example, a Libor-based loan could fall back to a term-RFR rate, while the derivative could fall back to a compounded overnight RFR rate, creating potential complexities for hedging and cash management.

Fallback language developed by the International Swaps and Derivatives Association (ISDA) does not include term RFRs, while fallback language proposed by the ARRC for loans does.

Eric Juzenas, a director in the Chatham Financial global regulatory solutions team, said that a hardwired approach has benefits over the long run because of the certainty and transparency it provides. Nearer term, however, it is difficult to predict how hardwired approaches will work in practice, because RFRs and spread adjustments are still being developed. Hence market participants must be wary of locking themselves into hardwired approaches that could lead to unpredictable results.

Flexibility for the Near Term

Chatham, which provides swap-related services to corporates, supported the amendment approach in its comment letter.

“Until different components of the transition waterfall - term rates, spread adjustments, etc. - actually exist and market participants have developed familiarity with them, then the hardwired approach becomes a risk if it doesn't function as effectively as wanted,” Juzenas explained. “It's the uncertainty in the interim that leads the buy side to favor a flexible amendment-based approach.”

A handful of lenders expressed support for “hardwired,” though with reservations. The Bank of Nova Scotia, for one, said that a more developed market understanding of the proposed fallback rates and the spread adjustment are necessary before adopting the specific documentation language for a hardwired approach.

“This is critical if we are asking borrowers to accept hardwired terms,” the bank says, adding, “For example, ARRC has proposed that forward-looking term SOFR be the primary fallback in the waterfall of replacement benchmarks, although this benchmark has yet to be developed.”

Efficiencies for Big Banks

Twenty-one comment letters were anonymous, and a majority of those favored the amendment approach. Juzenas said that large banks may support hardwired, even though a term RFR is not yet in sight, because they have sufficient resources to protect themselves should the hardwired approach experience glitches. Banks with thousands of loans on their books would find hardwired more efficient.

“Long term, a hardwired approach can benefit both borrowers and lenders. Now it's just how do you get there,” Juzenas said.

Comments on fallback language for floating-rate notes similarly homed in on the importance of developing forward-looking term rates. In a 2017 report, the Working Group on Sterling Risk-Free Reference Rates, which decided on the Sterling Overnight Index Average (SONIA) as an RFR, proposed two methods to calculate a forward-looking reference rate. One takes the fixed leg of a set-maturity, overnight-index-swap (OIS) contract (say, three- or six-month contracts) as the term reference rate, perhaps generated from executable quotes for OIS on regulated electronic trading platforms. Or, the term fixings could be derived from SONIA futures order-book data.

In a paper published in October, IBA describes a preliminary methodology to develop a SONIA term rate based on futures-contract data, as well as other approaches. It says that “the mathematics and the processes proposed to derive forward-looking rates for SONIA could be applied to other risk-free rates as markets and user appetite develop.”

SONIA in the Lead

SONIA is currently ahead of other RFRs in terms of its development. It was an existing rate that the market was familiar with, whereas SOFR is a new rate that the Federal Reserve Bank of New York began publishing last April.

ICE Futures - a separate entity from IBA within Intercontinental Exchange - and CME began trading SONIA and SOFR futures contracts last spring, and they have active markets in one-, three- and six-month contracts.

The first SOFR-based cleared swap was traded in October, and there have been several offerings of SONIA- and SOFR-based bonds and other loan products, mainly by governmental institutions and banks. On October 25, Toyota Motor Credit Corp. issued $500 million in three-month commercial paper priced off SOFR.

Active futures markets are lacking in other RFR benchmarks being developed in Japan, Switzerland and the Eurozone; in the case of the Eurozone the benchmark has yet to be launched.

Forward-Looking Readings

Transitioning cash markets is challenging because RFRs are overnight rates, and borrowers won't know what their full interest payments are until the end of the term, after compounding each day's rate in arrears. Futures and cleared swaps - another potential building block for term RFRs - are forward-looking derivative contracts for which there are highly transparent and liquid markets, and at least in theory they should provide strong reference points for term RFRs.

The IBA white paper aims to “provide a credible and representative forward-looking estimate for what RFRs such as SONIA and SOFR might average over one-, three- and six-month periods, so that can be imbedded into financial contacts,” said Timothy Bowler, president of the ICE affiliate. “That could give entities borrowing money the opportunity to reference a benchmark so they can have certainty around cash flows over those periods, as they do with Libor today.”

To that end, IBA has launched a Term RFR Portal where market participants can track current RFRs - now those of SONIA, SOFR and Japan's Tokyo Overnight Average (TONA), and one-, three- and six-month term rates, when they arrive. The portal also provides the interest compounded in arrears, enabling a borrower with a loan contract that references SOFR and pays quarterly based on the compounded interest over the last three months.

“When the bank sends the bill for interest due for SOFR over the last three months, and you want to check that number, you can go to our portal,” Bowler said.

He added that the harder step will be to provide forward-looking terms for RFRs. He said there is sufficient trading in SONIA futures contracts today to provide indicative forward looking rates for one-, three- and six-month terms, but not yet enough for them to be used as a reference benchmark.

“We would like to see a pick-up in activity in order to meet the requirements for an IOSCO- [International Organization of Securities Commissions] compliant benchmark,” Bowler said, noting that activity today is insufficient for a reference in financial contracts. “We think the market needs more time to develop, hopefully in the next year or so.”

The ARRC has stated that it plans to start publishing indicative term SOFR rates by early 2019.

Unresolved Issues

Cash products make up only about 5% of the $200 trillion in financial products referencing Libor, with the rest comprising derivatives that are mostly standardized and much easier to replace with RFR-based instruments. Nevertheless, those cash products provide much of the financing for corporates, floating-rate mortgages and other important markets. So the development of term RFR rates is a critical step to monitor in the transition away from Libor, but it's not the only one.

Chatham's Juzenas said that deriving the spread adjustment will be another key issue, and one impacted by the hardwire vs. amendment fallback issue. Since Libor and RFRs are determined in very different ways - Libor from ever fewer interbank lending quotes, and SOFR from, for example, $800 billion in overnight repurchase agreement transactions - their rate levels and behavior are also very different.

Because Libor is unsecured and SOFR is based on secured transactions, they are anticipated to react differently under stress. Consequently, generating a method to capture the differences between Libor and each of the RFRs is a major challenge that has yet to be resolved.

“The spread adjustment tries to capture not just the difference between the curves in a single point in time, but over a period of time as the curves change,” Juzenas said. “We understand how Libor acts over time, but we don't have as good a feeling for how compounded SOFR or term SOFR operates over time.”

Hardwired fallback language would require a set spread-adjustment methodology, but if that methodology proves faulty, there could emerge an unwanted transfer of value between parties in a transaction. The amendment approach, instead, would allow for adjustments to avoid such transfers.

Another significant issue that has yet to be addressed, Juzenas said, is the potential for mismatches between debt priced off a term RFR and the rate for swaps and other derivatives that market participants use to hedge their exposures, which so far only use the compounded overnight RFR. That mismatch is problematic for borrowers seeking to hedge floating-rate debt or other cash financial products.

Question two in the floating-rate-note consultation asks whether a forward-looking term rate should “be the primary fallback for floating-rate notes referencing Libor, even though derivatives are expected to reference overnight versions of SOFR.”

Importance of Consistency

Many of the comments to ARRC strongly support a forward-looking term but ignore the mismatch resulting from derivatives using the compounded overnight rate. That may be, as one commenter notes, because a minority of investors and issuers hedge the interest-rate risk of floating-rate note products, although addressing the “potential term-rate inconsistency would seem preferable.”

Another adamantly said that a forward-looking term rate should not be the primary fallback if it differs from the derivatives' version, since consistency “between the derivatives and securities markets is paramount.”

Another anonymous writer noted that the derivatives market should seek accommodation. “Our view is that if the cash markets prefer a term SOFR, ISDA should take steps to ensure the derivatives documentation is aligned,” the letter said.

The Farm Credit Banks' comment letter notes that sub-workgroups of ARRC and ISDA have tackled fallback issues in light of their respective constituents for some time, and their lack of coordination is problematic: “In our view this lack of coordination could create needless substantial financial basis risks to all financial institutions if, for example, triggers for different types of instruments are invoked at varying times.”