Replacing LIBOR in Derivatives Agreements

In July 2017, the CEO of the U.K. Financial Conduct Authority (FCA), Andrew Bailey, announced that the FCA will discontinue the London interbank offered rate (LIBOR) at the end of 2021. LIBOR is an interest rate index that is used in calculating floating or adjustable rates on trillions of dollars in bonds, loans, derivatives and other financial agreements.

A phase-out of LIBOR will be a major undertaking, and will raise many questions including regarding the existence of a successor rate for new transactions and what the fallback rate should be for existing transactions. Many of these questions remain unanswered for now.

Why LIBOR Is Being Replaced

The interbank offering rates (IBORs) are floating rates based on the actual or purported interbank offered rates for short-term loans in various currencies and maturities based on daily surveys of major banks. There are IBOR rates for each major currency (USD-LIBOR/GBP-LIBOR/EUR-EURIBOR/JPY-TIBOR). Following alleged manipulation of LIBOR after the financial crisis, the U.K. government decided to regulate the setting and administration of LIBOR, as well as the submission of rates by banks used to create LIBOR and placed these activities under the supervision of the FCA.

The alleged manipulation also prompted a review of major financial benchmarks and, in 2014, both the Financial Stability Board (FSB) and the Financial Stability Oversight Council (FSOC) in the U.K. reported concerns over the reliability and robustness of IBORs. The main issues raised by the FSB and the FSOC were that the rates banks report in order for these benchmark rates to be created do not have to be based on actual transactions, and overly rely on transactions in a relatively low-volume market, which increases the potential for manipulation. As a consequence, the FSB recommended that IBORs, and other similar benchmark rates, be determined to the greatest extent possible based on real transaction data. The FSB also recommended the development of alternative nearly risk-free reference rates (RFRs), because the counterparty risk of banks facing each other, which is accounted for in the IBORs, does not necessarily make sense for many of the transactions using the IBOR benchmarks.

In the U.S., following the FSB recommendations, the Federal Reserve assembled a group of financial institution representatives known as the Alternative Reference Rates Committee (ARRC) in order to identify alternative, transaction-based reference interest rates to replace USD-LIBOR. In June 2017, the ARRC announced that it had identified a new and currently unpublished overnight “broad Treasuries repo financing rate,” based on transaction-level data from certain tri-party and bilateral repo clearing platforms, as a potential successor to the USD-LIBOR (the Secured Overnight Funding Rate, or SOFR).

The following month, the FCA announced that despite its efforts to improve the setting of LIBOR, it had proven difficult to ensure that LIBOR rates and submissions were linked to actual transactions. The FCA concluded that it was unsustainable for market participants to indefinitely rely on reference rates that are not supported by an active underlying market, and announced that the LIBOR would be transitioned away by the end of 2021. It is important to note that there will not be a “ban” on LIBOR at the end of 2021, and in fact, LIBOR may still be published after that date by the LIBOR administrator (currently the ICE Benchmark Association), if it can obtain sufficient submissions from major dealers. There is no assurance that this will be the case. The FCA has the regulatory power to compel major dealers to provide submissions, although the dealers currently, if reluctantly, provide submissions on a voluntary basis. The FCA has announced that after 2021, it will not use its regulatory power to compel submissions, and given the environment, it is questionable whether the dealers will continue to voluntarily provide the submissions.

The FCA also noted in its announcement that alternative benchmarks should be established during the transition period in order to avoid disruptions and, as could be expected, attention in the U.S. turned to the potential alternative SOFR freshly announced by the ARRC.

It should also be noted that in the European Union, a new regulation on benchmarks in financial instruments/contracts will become effective on Jan. 1, 2018 (the EU Benchmark Regulation). Certain provisions, including rules relating to “critical benchmarks” (which are expected to include LIBOR), are already effective. One of these rules gives the relevant competent authority (in this case the FCA) the power to require the administrator of a critical benchmark to continue to provide the benchmark until it is transitioned to a new administrator. The relevant authority can also force dealers to commit to provide quotes to the benchmark administrator. However, under such powers, the relevant authority cannot impose these requirements on administrators and dealers for more than 24 months. The FCA and the relevant dealers nevertheless agreed to extend this period until the end of 2021.

Transition to a LIBOR Successor/Fallback Rate

The replacement of LIBOR has direct consequences for derivatives transactions, and in 2016 the International Swap and Derivatives Association (ISDA) established working groups on alternative risk-free rates and the development of fallbacks. Recently ISDA arranged a webcast to provide an update on the direction taken by its working groups. Although no consensus has yet emerged, ISDA has outlined the following approach:

• ISDA will recommend a fallback rate to replace the USD-LIBOR if it is permanently discontinued. ISDA confirmed that the current approach (which may still change) is to consider SOFR as the successor/fallback rate for USD-LIBOR.
• ISDA will amend the 2006 ISDA definitions in order to change the existing fallback mechanism if the reference rate in derivatives transactions is discontinued. This change, however, will apply only to derivatives transactions entered into after the date of the amendment.
• ISDA will develop a mechanism to change the fallback in existing transactions. ISDA typically facilitates this type of industrywide amendment by publishing a protocol amending all relevant derivatives transactions between two signatories of the protocol. During the ISDA webcast, it was mentioned that ISDA will most likely adopt the same approach in this instance.

Difficulties and Issues Raised by Market Participants

Buy-side market participants have raised multiple issues with the approach outlined above, most of which revolve around the choice of SOFR as a fallback rate and the risk of value transfer occurring upon the fallback provision being triggered.

Indeed, one of the main issues is the fact that SOFR is an overnight rate while the LIBOR rates have different rates for specific tenors. In addition, the LIBOR takes into account bank credit risk while SOFR is a risk-free rate. In order to avoid significant value transfer on the day the fallback is activated (due to these differences between the structures of LIBOR and SOFR), ISDA has proposed that the fallback will be publicly available as a screen rate and quotes will be based on tenors at one, three, six and 12 months (just like the LIBOR). In order to achieve this, the ISDA working group proposed that the fallback will consist of the SOFR plus a spread. The spread would be based on a snapshot of the LIBOR SOFR basis (i.e., the difference between the cash price and the futures price) across different maturities on the LIBOR's last day of existence. The spread would then be frozen from that point. After ISDA outlined this proposal, buy-side market participants were quick to point out that since the spread calculation will be based on a snapshot of the basis market on LIBOR’s final day, certain market participants could manipulate the spread by influencing the basis market during the previous days. One way to make market manipulation more difficult would be to base the fallback rate on a historical average, but certain market participant are concerned that a long-dated average would distort the rate. For now, there is no definitive solution to this problem, although lately the ISDA working group appears to be giving more consideration to a historical average approach in order to avoid market manipulation.

Another issue is the creation of a liquid market for SOFR swaps and LIBOR-SOFR basis swaps (which will be necessary to price the spread for different terms). The fact that the SOFR is still unpublished and should only be published in the first half of 2018 limits the work that can be done by the ISDA working group and by market participants to define a transition plan that will greatly depend on the liquidity of SOFR swaps and LIBOR-SOFR basis swaps. In addition, the historical data set available to determine a methodology for calculating the spread will be limited because the SOFR is a new index. Certain market participants have pointed out that it will be difficult to know how the spread will differ in case of market stress, especially because the LIBOR is an unsecured rate while the SOFR is a secured rate (which can create differences in volatility in case of market stress). In summary, only a little progress can be made until the SOFR is published, and the fact that it is a secured rate, and that little historical data will be available because SOFR is a new rate, will complicate the task of the ISDA working group in determining a fair spread for various maturities.

Finally, the fallback in the ISDA definitions will only be triggered if a “permanent discontinuance” of a benchmark rate occurs. The ISDA working groups in charge of this matter are creating a test that is based on the occurrence of the following events:

• Insolvency of an IBOR administrator without a successor being promptly appointed (within a specific time frame).
• A public statement from the administrator that it will cease publishing the IBOR, either permanently or indefinitely, without a successor being appointed.
• A public statement from the supervisor of an IBOR administrator that the IBOR has been permanently or indefinitely discontinued.
• A public and official statement from the supervisor to the IBOR administrator that the IBOR may no longer be used.

However, since the FCA is merely proposing to cease requiring dealers to make submissions (and not to prohibit the use of the IBORS), LIBOR could potentially continue to exist past the end of 2021 on the basis of fewer quotes. If it does, based on the above definition of “permanent discontinuance,” this would not cause LIBOR to be permanently discontinued and the ISDA fallback would not be triggered. This point is currently being discussed by the ISDA working group.

Conclusion

Although many issues remain unresolved at this time, a consensus for one or maybe multiple successor rates to LIBOR will likely emerge in the near future. The successor rate(s) may come from the work currently being conducted by ISDA, the Federal Reserve, or other industry or regulatory initiatives. Market participants are advised to monitor developments in the area, especially if they are contemplating entering into derivatives transactions maturing after the end of 2021.

Finally, it should also be noted that although ISDA working groups are working in the context of derivatives, the issues they are grappling with are also relevant to other financial agreements, and the work conducted by ISDA working groups could influence the selection of a consensus successor to LIBOR in this broader context as well.