From Libor to SOFR

By Barry Mills

There has been a lot in the financial press recently about Libor—the London interbank offered rate—and efforts to identify and implement alternatives to Libor. A measure of the rate at which large banks can borrow from one another on an unsecured basis, Libor is a reference rate underpinning more than $350 trillion in mortgages, commercial loans, bonds and derivatives. Reference rates serve an important purpose in that they reduce search costs and improve the ability of customers to compare pricing. With the publication of a reference rate such as Libor, bank customers are better able to judge whether the rate being charged is competitive.

In 2014, the Federal Reserve convened the Alternative Reference Rates Committee to identify a set of alternative reference interest rates that are more firmly based on actual transactions. Not only has Libor been subject to manipulation by market participants, but U.S. and other global regulators have become worried that with fewer banks borrowing in an unsecured fashion, the robustness and reliability of Libor is questionable. While significant improvements have been made to Libor, the U.K.-based Libor regulator has indicated that it will no longer utilize its regulatory authority to compel banks to submit to Libor. “Libor may remain viable well past 2021, but we do not think that market participants can safely assume that it will,” says Federal Reserve Board Chairman Jerome Powell.

How Libor came to be

In order to understand what the potential discontinuation of Libor and the adoption of a new rate might mean, it’s important to understand the genesis of Libor and how it became so prolific in financial contracts. Minos Zombanakis—a Greek-born banker who worked for U.S.-based banks in London—devised Libor in 1969 when he arranged an $80 million loan for the cash-strapped shah of Iran. Importantly, the loan was one of the first that could be split up among banks and charge a variable rate of interest to reflect changing market conditions. Loans in which multiple banks hold a share of a loan are called “syndicated loans.”

Zombanakis stated that he wasn’t looking to break new ground in the financial space but rather was simply trying to find a rate on which all of the syndicated loan participants could agree. At the time that Zombanakis created Libor, London was increasingly a global financial hub as Russia, China, and many Arab states chose to bank in Britain rather than the U.S. out of fear of confiscation. London also benefitted from U.S. regulations that effectively capped the amount of interest American banks could pay on deposits.

Zombanakis knew that no single firm would be willing to lend $80 million to a developing country, particularly at a rate that was fixed. U.K. interest rates at this time were 8 percent and inflation was on the rise. Banks generally are wary of lending at fixed rates for long periods of time when interest rates are expected to rise due to inflation. Variable rates loans help banks mitigate this risk since the interest rate payment that they receive fluctuates with prevailing market interest rates, which generally increase as inflation expectations increase. But how would Zombanakis decide on a variable rate upon which all banks that were involved in the syndicated loan could agree?

The solution that Zombanakis came up with was to charge borrowers an interest rate that would be recalculated every few months. The banks in the syndicate would report their unsecured funding costs right before the new rate adjustment period and this would serve as the basis for the variable interest rate charged to the customer. This way, changes to the interest rate received from customers on variable rate loans—with Libor as the reference rate—would roughly move in tandem with changes in rates associated with the syndicate banks’ own funding costs. With this invention, banks were more likely to participate in large syndicated loans so that their earnings associated with the syndicated loan wouldn’t be buffeted by changes in interest rates.

Before long, Libor became the basis for floating rate bonds and interest rate swaps, which allow companies to manage risk that arises from changing interest rates. Libor was adopted because it was perceived to be a simple and independent approximate measure of bank borrowing cost and, therefore, a candidate for a wide variety of variable rate loans. Indeed, the use of Libor in financial contracts of many types, including mortgages and student loans, proliferated.

Libor’s value for banks

A commercial bank making a longer-term variable rate loan while also funding itself with variable rate short-term borrowings will be affected by movements in both the general level of interest rates and the bank’s credit spreads (i.e., the cost of bank borrowing that is over and above general interest rates). If a bank were to use a Treasury note rate—generally perceived to be an approximation of a risk-free rate—as its reference rate in loan contracts, then the bank has only hedged the part of its funding costs that is associated with general movements in interest rates. If the spread at which the bank can fund itself widens relative to the risk-free rate serving as the reference rate in its lending contracts, then the bank’s net interest margin—or the difference between the interest earned on its loans and the interest that it pays on its borrowings—will suffer.

If instead the loan uses Libor as a reference rate, then the bank will also hedge market-wide bank credit risk. Being able to hedge market-wide bank credit spreads is important because, as was seen in the 2007-2009 crisis, Libor rates increased several percent while Treasury rates declined. The very fact that Libor could be used to hedge bank credit risk likely explains its expanded use in a variety of lending contexts.

While Libor’s use in loans is widespread, its prevalence in loans is vastly overshadowed by its use in interest rate derivatives. Indeed, interest rate derivatives account for nearly 90 percent of the outstanding gross notional value of financial products referencing Libor. While there is a strong case for the use of Libor for hedging a bank’s loan funding costs, it is unlikely that users of interest rate derivatives have an underlying motivation to hedge bank credit risk. Instead, users of interest rate derivatives tied to Libor prefer Libor as a means to hedge general interest rate risk.

Why then do users of interest rate derivatives rely on derivatives that are Libor-dependent if derivative users do not care about hedging bank credit risk? The answer is that the market for Libor-based derivatives is tremendously deep and liquid. The very fact that Libor is so prevalent in derivative contracts makes it that more challenging to switch to an alternative reference rate, even if that alternative would be preferable in that it does not contain bank credit risk.

If one were able to start from scratch, an ideal world would potentially involve two reference rates: one that would approximate to bank funding costs that could be used in bank loans and banking-oriented derivatives to hedge bank market-wide bank funding cost and a second reference rate that would be a riskless or near-riskless rate that would allow interest rate derivative-users the ability to more purely hedge movements in general interest rates without having to worry about movements in bank credit spreads.

SOFR and the path forward

As mentioned before, the continued publication of Libor is not guaranteed, and, therefore, Libor alternatives need to be considered. To effectuate implementation of a Libor alternative, one would need a deep and liquid market for the alternative rate or rates. Indeed, the criteria used by the ARRC when selecting an alternative to Libor for use in new derivative and other financial contracts included ensuring that the chosen rate has a deep underlying market that is robust over time. As Powell has said, “there would be no point in selecting a rate that might find itself quickly in the same kinds of conditions that Libor is in now.”

The ARRC has selected the Secured Overnight Funding Rate, a broad measure of overnight Treasury financing transactions as its recommended replacement for U.S. dollar Libor. The Federal Reserve will begin publication of SOFR in the second quarter of 2018. The ARRC has stated that the SOFR is the most robust rate available with underlying transactions of about $700 billion per day or more, much larger than the volumes associated with other potential Libor alternatives.

While the advantages of a robust reference rate are clear, SOFR has two distinct challenges relative to its suitability for lending arrangements. First, SOFR is a secured rate while Libor is an unsecured rate. This is important because Libor was originally developed to be and continues to serve as an approximation of a bank’s unsecured borrowing costs. Generally, secured borrowing rates are lower than unsecured rates because secured borrowings are backed by collateral. Thus, Libor will generally be higher than SOFR. Indeed, SOFR transactions involve Treasury securities as collateral. Second, SOFR is an overnight rate only, while U.S. dollar Libor is currently published according to the following tenors: overnight, one week, one month, two months, three months, six months and one year. Generally, longer-tenor instruments (e.g., one-year Libor) have higher borrowing costs than shorter-tenor instruments (e.g., an overnight rate). On the second challenge, the ARRC has announced that its paced transition plan includes the creation of a term reference rate based on SOFR derivatives once sufficient liquidity has been established.

Thus, the ARRC has come up with a potential solution—assuming sufficient liquidity is developed—to addressing the term structure challenge that SOFR poses, but the lack of bank credit risk in SOFR remains an unresolved issue. “Stakeholders in legacy loans may seek to include some form of a credit spread that incorporates the Libor-SOFR differential when Libor is discontinued into their interest rate calculations,” said the Loan Syndications and Trading Association in a comment letter.

The ARRC is also charged with devising plans for a voluntary transition that encourages the use of its recommended rate. One of the most complicated issues involves legacy contracts that reference Libor but do not have strong language in place if Libor were to discontinue publication. Initial reviews of loan documentation suggest that so-called fallback language regarding Libor alternatives—that is, how the contract addresses what happens if Libor is discontinued—may vary across institutions and products presenting challenges for industry-wide adoption of a Libor alternative. “While there may be no perfect contract language or fallback, good risk management requires that we work together to find language and fallbacks that are robust and that limit unintended valuation changes,” Powell says.

Next Steps

  • The work of the ARRC, which has to date encompassed recommending the SOFR and addressing preliminary implementation considerations for derivatives, will soon be summarized in a report.
  • The ARRC has announced that it will be reconstituted to more directly facilitate issues regarding contract robustness for loans, floating rate notes, and securitizations, as well as derivatives.
  • ABA is forming a working group to hold preliminary discussions regarding implementation considerations for loans, floating rate notes, and securitizations. Potential topics for discussion include the effects of a phased transition from Libor in legacy contracts and considerations for addressing Libor alternatives in new contracts. Working group participants may include loan officers, credit officers, in-house legal counsel, regulatory or compliance staff, risk managers, and financial function staff, among other potential interested roles. Please contact Barry Mills if you are interested in joining the working group or would like to learn more.