The End of LIBOR: Hotel California Edition (Part I)

Although July 3 was the first business day on which no USD LIBOR was published, many financial instruments will not reset until the next reset date, based upon the tenor of each instruments’ underlying benchmark, which could have been July 3, August 1, or some future date even six or 12 months later.

In our last installment, we provided a non-scientific analysis of variable rate indentures that do not fall under the purview of the Adjustable Interest Rate Act of 2022 (the LIBOR Act). Even for those bonds and other financial instruments, and their related swaps, that are covered by the LIBOR Act, the LIBOR transition will be anything but “smooth.”

SOFR – What’s in a Name?

The federal government enacted rules to establish default fallbacks for LIBOR-based financial instruments. As discussed below, these rules adopted the overnight Secured Overnight Financing Rate (Daily Simple SOFR) published by the Federal Reserve Bank of New York, and the proprietary “Term SOFR,” commonly referred to as “SOFR,” benchmarks administered by CME Group Benchmark Administration Ltd. (CME) as the replacements for LIBOR.

ARRC

These benchmarks were identified as alternatives to LIBOR by the Alternative Reference Rates Committee (ARRC). ARRC was formed in 2014 by the Federal Reserve and the New York Fed for the express purpose of finding a benchmark to replace LIBOR. ARRC primarily comprised of representatives from financial institutions and governmental entities.

ARRC’s final meeting was held on November 8, 2023.

Daily Simple SOFR

In June 2017, ARRC identified Daily Simple SOFR as representing the best practice for use in certain new US dollar derivatives and other financial contracts. Daily Simple SOFR began being published daily by the New York Fed on April 2, 2018.

Daily Simple SOFR is calculated by the volume-weighted median rate[1] for the repurchase market (i.e. the repo market), based on the cost for banks of borrowing cash overnight while posting US Treasury Bonds as collateral. It is also published utilizing 30-day, 90-day, and 180-day historical averages.

Daily Simple SOFR is based on a broad survey of actual transactions by the New York Fed involving hundreds of financial institutions. Daily Simple SOFR, which is an actual, historical benchmark rate, is considered to be a “risk-free rate.”

LIBOR
LIBOR had been based on estimated forward-looking rates aggregated by the British Bankers’ Association (BBA), a banking and financial sector trade association, from the subjective submissions of a panel of approximately 15 banks estimating at what rate they could borrow funds on the interbank market. Problems with this process included, among others, that bank submissions were not necessarily based upon actual transactions. This had occurred during periods of market turmoil when the interbank market froze (e.g. the Great Recession).

As a result of the LIBOR manipulation scandal, ICE Benchmark Administration Limited (IBA) assumed BBA’s role in aggregating LIBOR submissions in 2014.

LIBOR was considered not to be a “risk-free” rate.

Term SOFR

After numerous delays, ARRC formally recommended Term SOFR for use with commercial loans and most other financial products in August 2021.

However, ARRC did not support the use of Term SOFR for the majority of derivatives because, it asserted, those financial instruments generally already referenced Daily Simple SOFR and it was “essential to ensure financial stability” for the derivatives markets to transition to overnight risk-free rates.

CME publishes Term SOFR based on transactions in Daily Simple SOFR futures and benchmark Daily Simple SOFR derivatives to measure market expectations. Term SOFR’s forward-looking formulation makes it function as LIBOR did. Also like LIBOR, CME offers Term SOFR in one-month, three-month, six-month, and 12-month tenors, making Term SOFR an easier switch for contracts formerly set to a corresponding LIBOR tenor.

However, Term SOFR is not an actual rate. Instead, it is a benchmark rate determined on the futures markets. Importantly, it is not considered to be a “risk-free” rate.

In fact, immediately after the LIBOR end date, the International Organization of Securities Commissions (IOSCO) concluded that two credit sensitive rates and two Term SOFR rates had varying degrees of vulnerability in benchmark design, data sufficiency, and transparency. With respect to credit sensitive rates, IOSCO emphasized that market participants, primarily in the US markets, should proceed with caution if they are considering their use.

Likely due to these concerns as well as insignificant market acceptance, Bloomberg L.P. recently announced the termination of its LIBOR alternative – Bloomberg Short-Term Bank Yield Index (BSBY) – effective November 15, 2024.

Zombie LIBOR

Financial instruments that do not fall under US legal jurisdiction (and, thus, not subject to the LIBOR Act) have the additional option of adopting synthetic, or “Zombie,” LIBOR. As we discussed in a client alert earlier this year, the UK Financial Conduct Authority (FCA) concluded in April that maintaining synthetic LIBOR was necessary for legacy LIBOR contracts that would not be able to transition before LIBOR’s end date or were not subject to US law. Now that the LIBOR phase-out has occurred, the IBA plans to continue publishing one-, three-, and six-month LIBOR tenors using this new methodology and will continue to do so until September 30, 2024.

In the US, the Federal Reserve expressly disclaimed Zombie LIBOR as not being representative of the underlying market that the corresponding LIBOR rates were intended to measure. However, it just so happens that this “unrepresentative” calculation uses the relevant Term SOFR tenors published by CME plus the same fixed spread adjustment utilized in the LIBOR transition.

LIBOR Act

Rationale

As discussed in our previous alert, the LIBOR Act applies to “any contract, agreement, indenture… that, by its terms, continues in any way to use LIBOR as a Benchmark as of [June 30, 2023],” and officially set the end date of LIBOR as the first business day after July 1, 2023. The LIBOR Act was enacted to establish a clear and uniform process across the nation for replacing LIBOR in existing contracts to preclude litigation.

The LIBOR Act ultimately delegated the decision of what benchmark replacements should apply in connection with LIBOR’s transition to the Federal Reserve. It also provided that the Federal Reserve would promulgate regulations to carry out the LIBOR Act by no later than 180 days after its enactment, roughly September 11, 2022.

Implementation

Four months after this deadline, on January 26, 2023, the Federal Reserve published its final rule implementing the LIBOR Act. The Federal Reserve Rule, consistent with the LIBOR Act, utilizes replacements that are based upon Daily Simple SOFR and incorporates spread adjustments for each specified tenor of LIBOR. Notably, the Federal Reserve agreed with ARRC’s earlier observation that derivatives should not be based on Term SOFR subject to limited exceptions.

In contrast, Term SOFR was generally to be utilized for bonds, loans, and other non-swap financial instruments.

Implementation Risks

Basis Risk

The decision to utilize overnight Daily Simple SOFR for LIBOR swaps and one-month or three-month Term SOFR for other LIBOR-based financial instruments is problematic. To some degree, it defeats the essential purpose of interest rate swaps, which is to fully hedge the variable rate risks of a variable rate financial transaction.

To do this, swaps need to have the same benchmark as the underlying variable rate debt. If those benchmarks do not match, the issuers or borrowers are taking on additional risk known as ‘basis’ risk, which was not present when both the swap and the financial instrument were based upon the same tenor of LIBOR. In other words, the Federal Reserve Rule has now established a default scenario where the parties have taken on a new undisclosed risk.

Tenor Risk

As most bonds, loans and notes reset monthly or quarterly, one-month Term SOFR or three-month Term SOFR will be utilized in lieu of one-month LIBOR or three-month LIBOR.

However, as the Federal Reserve Rule strongly recommends that derivatives be based upon an overnight benchmark, Daily Simple SOFR, the related swap will not have the same tenor of the underlying financial instrument. Consequently, this will create additional potential risks for the issuer, borrower, and bank.

Viable Alternatives

While the LIBOR Act and the Federal Reserve Rule establish a default fallback when LIBOR ended, it did not make Daily Simple SOFR nor Term SOFR benchmarks mandatory for all financial instruments. Rather, the LIBOR Act and Federal Reserve Rule each permit parties to a financial instrument to select a benchmark replacement other than the applicable benchmark replacement set forth in the Federal Reserve Rule.

Of course, choosing an alternative LIBOR transition to eliminate basis risk and tenor risk, as well as to provide better legal and financial market disruption protections to issuers and borrowers, will be wholly dependent upon the understanding of these and other risks, and other business and legal financial ramifications associated with the LIBOR transition. As important will be the ability of issuers and borrowers to negotiate appropriate terms which, unfortunately, often depends upon the leverage these entities have with their lenders which may be lessened in today’s less favorable borrowing climate.

Upcoming Episode

Next time, we will examine constitutional and contractual issues arising from the LIBOR transition.

From a borrower and an issuer perspective, LIBOR, with a parallel swap, was meant to result in a cheaper costs of funds as compared to a fixed rate transaction, and provide less risk and more certainty in the event of significant interest rate fluctuations as we have recently encountered.

So too with Term SOFR:

There she stood in the doorway;
I heard the mission bell
And I was thinkin’ to myself,
‘This could be Heaven or this could be Hell’
- Eagles, “Hotel California”

[1] Volume-weighted median is calculated by taking all of the relevant transactions, ordering them from lowest to highest, summing their volumes, and then identifying the rate with trades at the 50th percentile of dollar volume.