In early April 2018, financial markets took one large step forward in the transition away from the London Interbank Offered Rate (LIBOR), the ubiquitous interest rate benchmark that is to be “phased out” by the end of 2021. On April 3rd, the Federal Reserve Bank of New York began publishing the highly-anticipated Secured Overnight Financing Rate (SOFR), which is the recommended replacement for LIBOR for U.S. Dollar denominated transactions (USD-LIBOR).
Why is this important? At present, LIBOR is referenced in trillions of dollars of financial contracts. Because of concerns about how LIBOR is calculated, the lack of liquidity in the interbank markets that are used to compute LIBOR, and the risk that LIBOR may be manipulated by the banks that submit pricing data to the ICE Benchmark Administrator (IBA) (see, e.g., the LIBOR scandal, and this fabulous book on the topic), the United Kingdom’s Financial Conduct Authority has determined that LIBOR should be phased out by 2021.
SOFR was chosen by the Alternative Reference Rate Committee (ARRC), a public-private initiative convened by the Federal Reserve in 2014, as the alternative to USD-LIBOR. In choosing SOFR, the ARRC considered a variety of factors, including: the rate’s usefulness to market participants, the depth of the underlying market, its likely robustness over time, and whether the rate would be consistent with the IOSCO Principles for Financial Benchmarks. The ARRC also set forth a paced transition plan for the adoption of SOFR, including a timeline that starts with creating the infrastructure for SOFR derivatives in the first half of 2018, and ends with the creation of a term reference rate based on SOFR derivatives at the end of 2021.
Here is what you need to know about SOFR. It is:
- An index that reflects a broad measure of the cost of borrowing cash overnight collateralized by U.S. Treasury securities (aka the “repo” market)
- Calculated based on transaction data, rather than subjective input
- Based on collateralized transactions, reflecting a nearly “risk-free” rate
- Comprised of data from an underlying liquid market (with about $800 billion in daily transactions)
Meanwhile, LIBOR is:
- An index that reflects the interest rates at which banks borrow from one another on an uncollateralized basis
- Calculated daily in 5 currencies (USD, EUR, GBP, JPY, CHF) and for maturities ranging from overnight to 1 year
- Formulated from pricing contributions by 17 panel banks (using “expert judgment” rather than transaction data)
- Comprised of pricing data from a far less liquid market (i.e., with only $500 million in daily transactions for 3-month USD-LIBOR in 2017)
As you can see, these two benchmarks are in no way economically equivalent to one another. Although SOFR addresses many of the liquidity and manipulation vulnerabilities of LIBOR, each benchmark measures different degrees of risk, and SOFR reflects a spot market while LIBOR reflects a term market. And because SOFR is collateralized and LIBOR is unsecured, one would expect the former to always track lower than the latter. Because of this discrepancy between the rates, if SOFR is substituted into a LIBOR-based transaction, this has the potential to create basis risk for the transaction participants. Many of these issues will need to be worked out by the market – creating a basis market between LIBOR and SOFR, and establishing SOFR derivative products to simulate a term market.
Along those lines, in keeping with the ARRC’s paced transition plan, the CME Group launched new SOFR futures contracts on May 7, 2018. On day one of trading, more than 50 entities traded about 3,000 one-month and three-month SOFR futures. This was a quiet start in comparison to the other rate futures contracts traded on CME’s exchanges, but in the early days trading volumes are expected to remain low in SOFR-based contracts, at least until market participants are more familiar with SOFR and the transition from LIBOR is well underway.
And now that we have a few weeks of SOFR under our belts, the market has started to weigh in with its initial reactions. For starters, the benchmark had a rocky opening, which some analysts say is to be expected – the overnight repo market is known to fluctuate, particularly with spikes at month-, quarter- and year-end, and dips mid-month. In addition, about two weeks after the rate was first published, there was a calculation error and certain data was inadvertently included in the Federal Reserve’s calculation of the rate. This error has since been addressed, but may be a setback in market adoption. Time will tell.
To complicate matters, while financial markets are in the process of phasing out LIBOR, the publisher of LIBOR (the IBA) recently announced that it is taking steps to make LIBOR a more “robust and sustainable benchmark.” Technically, it has every right to do so – while the UK Financial Conduct Authority has stated that market participants should transition away from LIBOR, there is no regulatory ban on continuing to publish or rely on LIBOR past 2021. Without an actual prohibition, the IBA has little incentive to stop publishing the ubiquitous benchmark – over $370 trillion of financial products are pegged to LIBOR, there is still market demand to use the benchmark, and the IBA is paid fees for licensing it out.
At present, the IBA is in the process of transitioning its panel banks to the new “waterfall methodology” for calculating LIBOR, and expects this changeover to be completed by early 2019. Under this new methodology, LIBOR will still reflect the average rate at which panel banks obtain wholesale unsecured funding, but the submissions from the panel banks will be subject to a three-level waterfall. If a bank has sufficient eligible transactions, it will submit a volume-weighted average price based on these transactions to the IBA. This is Level 1. If a bank does not have sufficient eligible transactions, it would move to Level 2, which means submitting transaction-derived pricing data. And if a bank does not have sufficient data to make a Level 1 or Level 2 submission, it will submit a more subjective rate – one at which it could fund itself, unsecured, in the wholesale funding market (Level 3).
The IBA expects to continue publishing LIBOR based on this new waterfall methodology (to be updated as needed), so long as panel banks are willing to continue submitting data.
Interested in learning more about how the LIBOR phase-out-that-is-not-technically-a-phase-out will impact your lending and risk management practices and documentation? Join Michael Best’s Alec Fraser and Cheryl Aaron in discussion with Melanie Wheeler, General Counsel and Chief Compliance Office of DerivativePath, in a webinar titled “A New Game in Town: SOFR to Replace LIBOR.” Registration for the May 23, 2018 event is free. Click here to learn more.