Secured Overnight Financing Rate in Loan Transactions

By: Jason Amster, Lexis Practice Advisor

This article describes the Secured Overnight Financing Rate (SOFR), a broad credit-risk measure that is a frontrunner to replace the London Interbank Offered Rate (LIBOR). The article addresses how SOFR is calculated, how it compares with LIBOR, and its advantages and disadvantages in loan transactions. These are important considerations in legacy deals and in new financings, as either could soon face the prospect of losing the loan market’s primary pricing mechanism.

Exit LIBOR

LIBOR (sometimes referred to as the Eurodollar Rate in credit agreements) is flexible and widely accepted, being available for maturities ranging from overnight to one year and is calculated in five currencies. It has long been the baseline pricing mechanism in loan agreements (and many other contractual arrangements for that matter). However, its future is uncertain.

As of February 1, 2014, the responsibility for overseeing and administering LIBOR passed from the British Bankers Association (BBA) to the ICE Benchmark Administration Limited (ICE) following the LIBOR manipulation scandal of 2012. ICE said that LIBOR will continue to be calculated in the same manner as it had been under the BBA to minimize the impact of this change on existing lenders and borrowers. Also following the scandal, banks themselves no longer wanted to report LIBOR, for fear of becoming embroiled in LIBOR-related trouble. The United Kingdom’s Financial Conduct Authority, the regulator overseeing LIBOR, said that it would no longer require banks to provide LIBOR estimates beginning in 2021. Many market participants have concluded that, at that time, LIBOR will cease to be the predominant interest rate benchmark. The question then was what would replace LIBOR as the reference rate in the $200 trillion in contracts that use LIBOR as of 2016. Of that, there are about $1.5 trillion in syndicated loans and $800 billion in non-syndicated loans that would need to be converted to a rate other than LIBOR (the derivatives market makes up about 95% of the outstanding gross notional value of all financial products referencing LIBOR).

Existing deals would generally default to the alternate base rate (ABR) or prime rate. These rates have been provided in the credit agreement as an alternative to LIBOR in instances in which, for example, banks cannot ascertain LIBOR or LIBOR does not accurately reflect their cost of funding. However, borrowers prefer to borrow at the LIBOR rate, which is lower. In fact, borrowers in default are generally prohibited from converting ABR loans to LIBOR. Therefore, simply switching over to these rates is not an ideal outcome for borrowers. For that reason, the loan market has begun to seek a viable alternative to LIBOR.

To address this problem and find a replacement for LIBOR, in 2014 the Board of Governors of the Federal Reserve System and the Federal Reserve Bank of New York (New York Fed) established the Alternative Reference Rate Committee (ARRC), comprising financial institutions and banks, trade associations (such as the Loan Syndication and Trading Association (LSTA)), and official sector members. ARRC noted that “the risks surrounding [LIBOR] pose a potential threat to the safety and soundness of individual financial institutions and to financial stability.” The ARRC set out to find an alternative reference rate to LIBOR, best practices for contract robustness, and plans to adopt and implement an alternative rate. It initially recommended as an alternative rate the Broad Treasuries Financing Rate, which subsequently became known as SOFR.

Enter SOFR

SOFR is based on several risk measurements for the purchase and resale of U.S. Treasury securities under repurchase agreements (as described below). As a secured rate, it cannot replace the unsecured LIBOR directly, as the SOFR rate on any day is lower than LIBOR. Borrowers and lenders would have to determine an appropriate conversion mechanism.

The New York Fed began publishing quotes of the SOFR rate in April 2018. SOFR measures the cost of borrowing cash overnight backed by U.S. Treasury securities as collateral. It is a benchmark rate that incorporates trading data from three risk-free reference overnight repurchase (repo) rates. In a repo agreement, a dealer (or borrower) sells a government security to investors and buys it back at an agreed-on higher price at a later date (in this case, the next day). The difference between the selling price and the repurchase price (i.e., the discount) is the basis of the repo rate and is the same as an interest rate. Treasuries can be traded through repos in three ways:

• Tri-party repo, which uses a clearing bank as a go-between for a specific buyer and seller
• General Collateral Financing (GCF) repos, which are like tri-party repos but are traded on exchanges, with transactions between anonymous buyers and sellers (i.e., they are blind brokered) settled on the clearing banks’ platforms
• Bilateral repos, which are direct transactions between buyers and sellers. These do not use third parties such as clearing houses, but they may be cleared through the DeliveryVersus-Payment (DVP) service offered by the Fixed Income Clearing Corporation (FICC)

The New York Fed has included all three of these types of repo transactions in its calculation of SOFR. SOFR is calculated as a volume-weighted median of transaction-level tri-party repo data collected from the Bank of New York Mellon as well as transaction data and data on bilateral U.S. Treasury repo transactions cleared through DVP, which are obtained from Depository Trust & Clearing Corporation (DTCC) affiliate DTCC Solutions LLC. The New York Fed also includes in SOFR data from its Broad General Collateral Rate (a tri-party GCF rate). The FICC acts as a central counterparty for GCF and bilateral repos. Removed from these calculations are transactions the New York Fed refers to as “specials,” or specific-issue collateral. These specials trade at cash-lending rates (i.e., rates lower than those for general collateral repos); cash providers accept this lower yield, so they can obtain a particular security.

The New York Fed publishes the result of this calculation, SOFR, on its website at about 8 a.m. New York time each business day.

Differences between LIBOR and SOFR

SOFR differs from LIBOR in that there is a far higher volume of SOFR-based trading and in the underlying nature of the rate itself. Higher volume generally makes SOFR safer from manipulation than LIBOR. For example, just after the New York Fed began quoting the SOFR rate, $754 billion in daily trading volume made up SOFR, as opposed to $500 million in three-month LIBOR, according to the ARRC. Thus, a very large number of contracts based on LIBOR (see below) are derived from a relatively small amount of underlying trades—making LIBOR susceptible to manipulation. The relative weakness of LIBOR was exacerbated by the financial crisis and subsequent LIBOR scandal, when banks drastically limited (in fact, nearly eliminated) reporting LIBOR. SOFR is based on data from far more trades than is LIBOR, ideally making it a more accurate measure of the cost of credit.

LIBOR and SOFR themselves are based on different metrics as well. The most significant difference between LIBOR and SOFR is that LIBOR is an unsecured rate and represents banks’ estimates as to their cost of funds. SOFR meanwhile is a secured, risk-free rate. Thus, LIBOR arguably reflects banks’ costs of funding more accurately than SOFR. On the other hand, the calculation of LIBOR was opaque, based on polling of certain banks. The calculation of SOFR is more transparent, based on market data. In addition, because of the size of the SOFR market and the different components that go into its calculation (see above), the ARRC concluded that SOFR does reflect the economic cost of lending and borrowing relevant to a wide array of market participants.

In addition, LIBOR quotes are available for deposits with several different maturities (interest periods), from overnight to one year. At its launch in April 2018, SOFR lacked a term reference rate, being limited only to an overnight rate (the ARRC was unable to find a term rate like LIBOR that otherwise met its criteria for a replacement rate). However, in May 2018, the ARRC published an indicative three-month SOFR rate. Otherwise, issuers selling bonds tied to SOFR have been making do with the overnight rate (Fannie Mae made the first issuance of SOFR bonds, $6 billion worth in July 2018). That is, a term interest rate is derived from extrapolating from the daily SOFR rate (i.e., an average daily rate). The disadvantage of such extrapolation is that the parties do not know the final rate at the start of the interest period, as is the case with LIBOR. However, SOFR has begun trading on futures markets, and this should allow for the calculation of true term rates. The ARRC timeline anticipates the creation of a SOFR term reference rate as the final step in its “paced transition plan,” expected to be completed by the end of 2021.

The differences between LIBOR and SOFR are most essentially represented in the underlying rates themselves. For that reason, you cannot simply amend a credit agreement to replace LIBOR with SOFR. For example, on January 3, 2018, the overnight LIBOR rate was 2.39188% and SOFR was 2.7%. At other points in the cycle, SOFR has been lower than LIBOR, and the overnight rate for LIBOR is not the most representative of LIBOR measurements. In any event, some credit adjustments to the spread must be made to account for the difference rates.

Challenges and Next Steps

Practitioners should keep abreast of these changes, as they will have an impact on many existing credit agreements and on new deals. To avoid defaulting to costly ABR or prime rates, credit agreements should either have language allowing for a streamlined amendment to a to-be-determined benchmark (in this case, SOFR). Ideally, your credit agreement already has such a mechanism in place. Before you advise your client on this point, check to see what level of lender consent is necessary to make such a change. Most recently negotiated deals do have such language, and the parties can replace LIBOR through an amendment that can be blocked only by a required lender vote (i.e., a negative consent). Some credit agreements require no consent other than the administrative agent and the borrower, and an unfortunate few demand an affirmative required lender vote to make such a change.

New loans should have this LIBOR successor language baked into the credit agreement. Some things to consider in drafting these provisions are making the appropriate adjustments for the difference in spread between LIBOR and its replacement, what events will trigger a move to the new rate, how and who selects the new rate (e.g., the administrative agent alone, or the agent with borrower consent), and whether required lenders will be allowed to sign off on this change.

Jason Amster is a Content Manager for Lexis Practice Advisor®. Before joining LexisNexis, Jason was an associate in the Banking and Finance Group of Kramer Levin Naftalis & Frankel LLP, where he represented lenders and borrowers in asset-based and cash-flow lending, acquisition financings, and CLO-secured loan transactions. He also represented creditors and debtors in restructurings and work-outs, including debtor-in-possession and exit financings. Previously, he was an associate in the Banking and Credit Group at Simpson Thacher & Bartlett LLP. There, he worked on U.S. and cross-border transactions in connection with acquisition financing and project development.

To find this article in Lexis Practice Advisor, follow this research path:

RESEARCH PATH: : Finance > The Credit Agreement > Credit Agreement Guide > Practice Notes

Related Content