Just like the nail-biting countdowns of yore (Y2K anyone?) there is a new date to be skittish about: Dec. 31, 2021. After that date, the London Inter Bank Offered Rate (LIBOR)--one of the more popular reference rates to calculate interest payments on a host of financial instruments--will likely disappear. The clock is ticking, and with roughly $400 trillion of financial contracts using LIBOR (cumulative of all major currencies) currently outstanding, the stakes associated with this transition are high.

Working groups across regions have formed, and although many countries have already determined the planned replacement for their local interbank offering rate (IBOR), a slew of issues remain.

We believe institutions we rate will need to prepare their transition for the end of LIBOR over the next year and a half. In the Q&A below, S&P Global Ratings explains the dynamics of replacing LIBOR, including the operational and credit issues that may arise during the transition.

Frequently Asked Questions

What is LIBOR, why is it going away after 2021, and and what challenges does that create for financial market players?

LIBOR measures the average cost that banks pay to borrow from each other on an unsecured basis for particular short-term periods. It is measured in five currencies and seven maturities, from overnight to 12 months. A subset of a panel of 20 banks reports their unsecured interbank funding costs for each of those currencies and maturities, and a methodology is applied to average those submissions.

Along with a number of other IBORs--such as the Euro Interbank Offering Rate (EURIBOR) and Tokyo Interbank Offering Rate (TIBOR)--LIBOR is commonly used as a benchmark or reference rate for trillions of dollars of financial products and contracts. Despite the prevalence of IBORs, regulators and some other industry participants have been looking to move away from these reference rates, largely because of long-running concerns about data integrity and validity. For example, a LIBOR quote is not based on an actual transaction, but instead on a bank's self-reported funding cost measures. Some banks have manipulated IBORs--notably LIBOR during the financial crisis--through false reporting, so as not to expose possible credit issues with the bank. Several banks paid large fines for that manipulation, making it less attractive for them to continue participating on the LIBOR panel. In addition, since the financial crisis and the onset of quantitative easing, liquidity in unsecured interbank markets has declined, making IBORs less meaningful benchmarks. That liquidity can also dry up further in stressed times, causing significant instability in IBORs.

In 2014, the Financial Stability Board (FSB)--an international body made up by central bankers and regulators--published a paper recommending strengthening existing benchmarks and developing alternative benchmarks. Regulators have been trying to execute on those recommendations since then, most notably by seeking to replace LIBOR and several other IBORs with other, and sometimes new, benchmarks that meet certain principles of their design, governance, data sufficiency, and transparency. Regulators believe the replacement benchmarks should be based on observable, independent transactions (rather than self-reported inputs) in active markets with prices driven by supply and demand.

Is there any chance LIBOR will exist beyond 2021?

In 2017, Andrew Bailey, the head of the U.K.'s Financial Conduct Authority (FCA), which regulates LIBOR, agreed with the panel banks' request to sustain LIBOR until the end of 2021, giving the industry time to plan for transitions to alternative reference rates. After 2021, the FCA cannot ensure LIBOR quotes will continue to be published. Still, we believe the FCA could convince the 20 banks to continue to quote LIBOR for some extended period, if the transition to a new reference rate is not on track by 2021. But the banks would be under no legal obligation to do so.

What is the trigger for LIBOR's end?

The end of 2021 does not automatically mean an end to LIBOR. The trigger is expected to come from the FCA, tasked with monitoring the viability of maintaining LIBOR as a valid reference rate based on the number of submissions. We expect the FCA will most likely trigger the end of LIBOR at the end of 2021, but the regulator may offer an extension. In any event, the fewer banks that provide a LIBOR quote, the more likely the FCA is to determine that LIBOR is no longer a valid rate. Still, we believe market participants could privately call banks to get an unofficial LIBOR rate (assuming some banks would still provide it) to fulfill the legality of their contracts.

How could the change from LIBOR to a new reference rate affect rated banks?

Our base-case expectation is that the entities we rate will accelerate their preparation, particularly in 2020, and eventually make adjustments that mitigate the risks associated with the end of LIBOR and other IBORs. However, as we move closer to year-end 2021, we will look closely at the readiness of individual entities, and we could reflect any perceived lack of preparation in our financial institution ratings.

After 2021, financial institutions that fail to prepare could have asset or liability issues, because existing contracts may become tied to an unexpected rate that weakens asset-liability management and profitability. A host of operational challenges and legal disputes could arise if:

• Banks are slow or neglect to set up adequate operational procedures to prepare for the end of LIBOR. This includes:
• --reviewing the current fallback language in existing contracts;
• --determining the fallback language they would prefer to switch the legacy contracts to (and determine language for new contracts written);
• --contacting all related parties to start the process of inserting this fallback language in these contracts and ensuring that new contracts contain proper fallback language
• Banks do not add a proper credit adjustment to the loans they generate based on alternative benchmarks to maintain adequate profitability margins and asset-liability management thresholds.

Which banks are most exposed to transitioning from LIBOR?

We believe the factors that may increase or decrease a banks' exposure to complications from a LIBOR transition include:

• The proportion of a banks' loans and assets linked to LIBOR
• The percentage of LIBOR exposures that remain outstanding after 2021
• The percentage of exposure that includes new fallback language anticipating the end of LIBOR
• The type of fallback language included, and whether it covers the risk of asymmetry that may arise from the use of an alternative rate

Will S&P Global rate debt tied to reference rates that replace LIBOR?

We have already rated instruments tied to Secured Overnight Financing Rate (SOFR) and the Sterling Overnight Interbank Average Rate (SONIA) and plan to rate Euro Short-Term Rate issuance as well. In order for us to rate an instrument tied to a variable-rate index, the index must meet certain minimum standards:

• Generally, the index must have an established track record or be considered by S&P Global Ratings as a successor to a long-standing index;
• The index must be readily accessible, such as posted on a public website;
• The index must be independent (for example, calculated by a third party independent of the issuer); and
• The index must be calculated in a transparent, consistent, and verifiable manner.

We expect the indices chosen by the various working groups to meet these characteristics.

How are regulators and industry participants preparing for the end to LIBOR? What reference rates are most likely to replace LIBOR on floating-rate instruments and derivatives?

Central banks and regulators around the world have organized industry working groups to make recommendations for transitioning from LIBOR and other IBORs to other reference rates (see table 1). In each jurisdiction, some of the challenges have been:

• Determining what reference rate could replace the IBOR;
• Encouraging market participants to determine how that rate could be adopted for various products and building a track record and liquidity in products (principally loans, bonds, and derivatives) tied to it;
• Establishing and implementing fairly standardized fallback language on newly issued IBOR-tied products that determines how a separate reference rate will replace an IBOR if phased out;
• Establishing fairly standardized fallback language that counterparties can use to amend existing IBOR-tied products that were not originally structured to anticipate a permanent end to that IBOR; and
• Determining what tax, accounting, and regulatory consequences the transition from an IBOR to a new reference rate could have and working with local tax, accounting, and other regulatory agencies to try to ameliorate any significant difficulties.

Table 1

The replacements differ from one another, both in terms of track record of existence (SONIA has the longest) and the transactions they're based on (whether secured or unsecured). The latter will add to the complexities of the transition, because each of these rates will react differently to market stress.

In some jurisdictions the transition to LIBOR may be less complex. For example, in Australia, the transition away from LIBOR is mostly a wholesale market problem and confined to wholesale market issuances and derivatives. Corporate and bank issuances offshore up until recently used LIBOR, and this needs to be addressed. More problematic, though, is cross-currency swaps, nearly all of which use LIBOR, some over very long time horizons.

What is the dollar amount of instruments outstanding that use U.S. dollar LIBOR as a reference rate, and what kind of instruments are they?

According to the Financial Stability Board Market Participant Group, the gross national value of all financial products tied to U.S. dollar LIBOR is about $200 trillion, roughly 10 times U.S. GDP, and $400 trillion including all major currencies. (There are also hundreds of trillions of dollars collectively that use other IBORs, particularly EURIBOR). The notional value of the derivative market accounts for 95% of all financial products that use U.S. dollar LIBOR as a reference. Positively, roughly 82% of contracts using U.S. dollar LIBOR will expire by the end of 2021, according to the second report published in 2018 by the Alternative Reference Rate Committee (ARCC), an group established to guide the U.S. transition away from LIBOR. This report is based on data from 2016. Still, this leaves about 18%, or $36 trillion of financial contracts, set to expire after 2021. In addition, a lower percentage of loans and securitizations tied to LIBOR (as opposed to derivatives) mature before the end of 2021. For instance, less than 60% of retail mortgages and mortgage-backed securitizations mature by then.

Table 2

What happens to currently outstanding financial instruments tied to LIBOR that are scheduled to mature after 2021?

The fallback language in the contracts that remain outstanding after 2021 governs what will happen to the instrument. The situation is complex, because the fallback language differs depending on the type of financial instrument (business loan, consumer loan, securitization, etc.) and can be unique to an individual product. That said, many contracts have a fallback waterfall that starts with instructions to seek quotes from a set of reference banks in the event LIBOR is not published. If these quotes cannot be obtained, then the fallback options vary, depending on the type of financial instrument (see chart 1).

The fallback language is general and even within a financial contract class could differ. Some fallback language refers to a conversion to the last LIBOR rate quoted, essentially turning a variable payment into a fixed payment. Other language may convert the payment to the prime rate--typically higher than LIBOR--resulting in an unplanned increase in borrowing costs.

Complicating the matter further is that some contracts don't contain fallback language, having never considered the disappearance of a LIBOR rate as a possibility. For these instruments, if an alternative rate is not agreed upon before LIBOR's disappearance, there could be litigation from investors and market disruption.

Despite the existence of previously written fallback language in legacy contracts, the parties that wrote such language probably only anticipated a temporary cessation of LIBOR rather than a permanent one. The parties involved now have an incentive to replace this language with new fallback language, so the replacement rate will be more in line with current expected payments.

To address the challenges of determining the right credit spread for a revision to a SOFR-based rate, some newly written contracts are using the amendment approach, basically saying that the reference rate will fall to the industry standard, to be determined at a later date. Presumably the industry standard will be whatever the working group in each jurisdiction is proposing (e.g., SOFR).

Which party in an existing contract has the ability to change the current fallback language?

The protocol typically depends on the type of instrument (derivative, loan, securitization, etc.) that the contract pertains to. Different instruments require different kinds of consent (see chart 2).

For derivative contracts that extend past 2021 and are traded on exchanges, consent is in the hands of the International Swaps and Derivatives Association (ISDA). ISDA will provide a protocol procedure, including the reference-rate language that it plans to publish by year-end 2019. ISDA has already identified appropriate conditions that should trigger a fallback and has identified SOFR (and other alternative rates selected by global working groups) as the appropriate fallback rate. Determining appropriate methods for calculating a credit component and term spread or any necessary adjustments to the new reference rate is ongoing. Positively, ISDA's work on LIBOR and other IBORs should set an example that could be used for cash products (nonderivative financial instruments) and over-the-counter bilateral derivative contracts.

We believe that cash products' transition from LIBOR will be more challenging than derivatives', as there is no ruling body, such as ISDA, to determine the reference rate, credit, and term components. For cash instruments, consent needs to be issued, most of the time unanimously, even for multilateral agreements such as shared national credit loans. This will prove difficult as there will likely be winners (those that receive a higher rate than current LIBOR contracts) and losers (those that receive a lower rate) depending on how the chosen reference rate is applied. Consumer loans can be changed by the noteholder only--perhaps a silver lining in terms of consent.

Besides determining appropriate replacement language, what are the other main issues that affect the transition out of LIBOR?

Other key issues with converting a LIBOR-referenced note to SOFR are as follows (see Appendix for more detail):

• The lack of term rates in SOFR, which may not be in place until shortly before LIBOR phases out
• The lack of a credit component in SOFR
• Technical issues that could arise in the repurchase (repo) market (supply/demand) that can cause spikes in repo borrowing rates, similar to what occurred in mid-September 2019
• Possible mismatching of interest payments
• Third-party vendor challenges

Are LIBOR-tied instruments still being originated? Do they contain contractual fallback provisions for how interest will be calculated post LIBOR?

Clearly, concern about the end of LIBOR has not hindered the origination of loans, bonds, securitizations, and derivatives tied to it. Originators of many products have either not begun or only recently begun to update fallback language. For instance, even earlier this year, we saw LIBOR-tied bonds with very long maturities (e.g., 40 years), with fallback language essentially unchanged from prior years.

In the residential mortgage market, the Mortgage Bankers Association in the U.S. has published a disclosure template for originators to share with borrowers on adjustable-rate mortgages. The template makes very clear that the end of LIBOR could affect the borrower's payment. However, it doesn't appear that the actual fallback language in these newly originated mortgages has been significantly updated (a process Fannie Mae and Freddie Mac are working on).

Positively, the recommended fallback language issued by ARRC in April 2019 seems to be having an impact, at least on bond issuances. Some recent bond issuances have contained language stipulating how SOFR would replace LIBOR in the event the latter no longer exists. We expect similar fallback language to become more widely adopted on issuances of bonds and other products over the next year.

How are regulators or legislators trying to ease the transition from LIBOR?

Given the number of contracts outstanding, the ARCC and the Fed reportedly are considering turning to New York's state legislature to ease the transition, because many notes outstanding are governed by New York law. Legislation theoretically could designate SOFR, or some variation of SOFR, as an acceptable substitute to LIBOR and possibly allay contract squabbles. Still, even if the New York state legislature gets involved, impairing the obligation of an existing contract will likely turn into a thorny legal matter. Even if legislation is passed, it could still be challenged in the courts, because choosing an alternative to LIBOR will inevitably favor one party in a transaction over another. In addition, although most commercial contracts are governed by New York state law, many consumer contracts are not, meaning products such as adjustable-rate mortgages and student loans would be out of scope.

Other governing bodies are engaged in the LIBOR transition process. ISDA is putting together protocols for existing and newly written derivative contracts. Other central governing bodies such as the Basel Committee on Banking Supervision and the International Organization of Securities Commission have stated that amendments to derivative contracts in response to LIBOR should not trigger new initial margin requirements. However, this has not yet been written into law, and it is unclear how narrowly it will be interpreted and enforced. In addition, the ARRC is working with the Commodity Futures Trade Commission and other regulators to ensure that the uncleared margin rule for derivatives would apply to a SOFR-based index

The Federal Housing Finance Administration, the regulator for Fannie Mae and Freddie Mac, has also said that at some point those entities will completely stop purchasing loans tied to LIBOR. Fannie Mae and Freddie Mac plan to create a SOFR-indexed adjustable-rate mortgage product. We expect loan originators of nonconventional products (i.e. mortgages not sold to Fannie Mae or Freddie Mac) will likely follow suit.

Could there be tax or accounting considerations as a result of transitioning to a new reference rate?

Switching from a LIBOR-based note to another reference rate will create sudden gains or losses on existing contracts because of a change in spreads. The ARRC is working with the IRS and Treasury to help nullify any tax gains resulting from incorporating a new reference rate.

There are also accounting consequences to modifying existing contracts with a new reference rate, such as hedging relationships. The ARRC is working with the Financial Accounting Standards Board and the SEC to ensure that a SOFR-based index would qualify for relief for hedge accounting. The SEC is also ensuring that financial documents properly disclose the steps an institution took during the LIBOR transition.

The world didn't end on Y2K, and financial markets most likely won't combust on Dec. 31, 2021, either. But transitioning from LIBOR won't be easy, and institutions that don't prepare for the event properly may run into issues, which if extensive enough, could lead to lower ratings.

Appendix

Details regarding the hurdles to switch from LIBOR to a SOFR-based reference rate

Term rates

SOFR is an overnight rate that changes daily, while LIBOR is quoted in seven maturities. The rate paid on most LIBOR-tied loans is set based on the value of LIBOR at the beginning of the interest-rate period. In that way, the borrower pays a rate equal to essentially what it costs banks to borrow from each other for that period (LIBOR) plus a spread. Therefore, there is a strong correlation between the rate the bank earns on the loan and that rate it pays to borrow.

Pricing a loan strictly on overnight SOFR on any given day carries greater asset-liability management risk. If SOFR moved significantly after that day, the bank's net earnings on the loan, considering its own funding costs, could rise or fall significantly. One solution to this is to use the daily SOFR rate--either averaged or compounded based on the 30 days at the end of the interest period or the 30 days prior to that period--to create a term rate. With the former, the lender would not know the yield earned until the end of the period. In either case, the bank's funding costs during the 30-day period could rise or fall significantly compared with the yield earned.

A forward-looking SOFR rate curve, based on derivative transactions, could ameliorate this. However, that can't happen until SOFR-based derivative contracts trade with significant liquidity, which probably won't occur until 2021. In other words, any bank looking to amend contracts based on SOFR forward rates rather than overnight average or compounded rates probably will not be able to do so until shortly before LIBOR phases out.

Credit

LIBOR is an unsecured rate that partly reflects the credit risk of lending to banks, while SOFR is a nearly risk-free rate based on secured borrowing transactions. This means that, in times of market stress, SOFR is likely to compress, while LIBOR spikes. Linking assets to LIBOR helps banks manage their interest rate risk, because higher payments on their LIBOR-based assets can offset the pressure of increased funding costs. With SOFR, issuers may need additional compensation via higher spreads for bearing the risk that during a stress period their asset yields could compress at the same time their funding costs rise. Perhaps, though, this is not an insurmountable problem since historically periods of significant widening between LIBOR and risk-free rates have been fairly short.

Chart 3

Technical issues with the repo market

Overnight repo prices can unexpectedly spike because of technical issues, even though credit is not a concern in the market. For example, banks may need to increase liquidity at the end of a quarter, pushing repo rates higher. Another example is the dislocation that occurred in mid-September 2019, when overnight repo rates surged for a few days because of a confluence of events including a surge in demand for cash, without enough supply of to satisfy this demand.

Mismatch of payments

Financial parties also need to be cognizant of a possible mismatch of interest payments. For example:

• The fact that there is varying fallback language in contracts could result in applying different reference rates in a host of contracts.
• An asset could switch to a floating-rate SOFR reference rate, while a liability could switch to a fixed-rate payment (for example, if the parties can't agree on a new reference rate, and the fallback language calls for switching to the last LIBOR rate quoted).
• The fallback to a new reference rate could occur at different times for different contracts. For example, there may be precessation triggers to end the referencing of LIBOR in one contract but not in another.
• The fallback language on one contract could revert to a U.S. dollar-based LIBOR replacement, such as SOFR, while another contract could fall back to another replacement rate, such as SONIA. The two reference rates will likely not behave in tandem in all economic conditions.

Third-party vendors

Most of the notes are currently paid either in-house or via a third party, whose systems may only be able to handle specific reference rates, such as LIBOR or prime. In order to make the proper computations of a SOFR-based rate, systems will need to be updated. If the third-party vendor doesn't know the calculation of the new rate until near the transition date, many glitches could arise and a host of legal issues could ensue.

Related Research

• The Secured Overnight Financing Rate (SOFR) Is Consistent With Our Principal Stability Fund Ratings Criteria, July 30, 2018
• With A LIBOR Phase-Out Likely After 2021, How Will Structured Finance Ratings Be Affected?, Oct. 19, 2017
• Credit FAQ: SONIA As An Alternative To LIBOR in U.K. Structured Finance Transactions, Feb. 6, 2019

Related Criteria

Principles For Rating Debt Issues Based On Imputed Promises, Dec. 19, 2014