LIBOR Transition End Game
By Joe Koltisko
Risks & Rewards, September 2021
It happens often enough in track and field: with one lap to go, a few upstarts pull away from the pack and make their bid to overtake the front runner who, despite heavy sponsorship and media coverage, has not put quite enough distance between them. That front runner is SOFR, the Secured Overnight Financing Rate, the designated successor to LIBOR. The sprint is on—and the odds are in its favor, in one of three avatars: overnight compounded-in-arrears SOFR, Simple Interest SOFR, and Term SOFR. Three risk-free reference rates and one index to bind them.
The upstarts to watch are: BSBY (rhymes with “frisbee,” which stands for the Bloomberg Short-Term Bank Yield Index); CRITR (Credit-Inclusive Term Rate) and CRITS (Credit-Inclusive Term Spread) from Markit Partners, the global economic and financial index data provider; the IBA Bank Yield Index—from the folks who bring you LIBOR, the ICE (Intercontinental Exchange) Benchmark Administration (IBA); and AMERIBOR, an index of exchange-traded debt issued by small banks.
In case you have been living in a cave for the last few years, or passing the pandemic confinement by binge-watching “Vikings” and “Lord of the Rings,” you may wonder what all the fuss is about. You probably also are keenly aware that dreams become reality, though not everyone’s dreams come true; good eventually triumphs over evil, and fate overrides the best-laid plans. TONAR is not one of Tolkien’s elves, it is the new risk-free rate in Japan. SARON is the Swiss risk-free rate, not to be confused with Sauron, Lord of Mordor.
LIBOR is going away because short-term bank funding transactions (never all that robust a data set) are not available in sufficient quantity to derive it. Instead, banks submit daily estimates of their funding costs (along with any real transaction data) to the IBA which uses the data to publish the rate. The estimates are fine most of the time, but when funding markets are dislocated they are pure fantasy. We got a taste of that in the market dislocation of March 2020 when the COVID lockdown restrictions were first announced, which redoubled the resolve of the regulators to find a better alternative.
From the point of view of the Federal Reserve, the Commodity Futures Trading Commission (CFTC), the Securities Exchange Commission (SEC), the Bank of England, and the UK Financial Conduct Authority (FCA) (the latter regulates LIBOR production)—LIBOR’s days have been numbered for the last 10 years. A good index is one based entirely on real transaction data. Policing the manipulation of LIBOR quotes is messy since it requires herculean, draconian controls on communications within and among market makers. It requires tedious and expensive enforcement of traders’ codes of honor. Ultimately it does not seem feasible to ask market makers to both “put your fear and greed engine to work to arb out inconsistencies you see in the pricing” and “be perfect saints in dealing with the buy side—ignore your own interests in providing advice and data to the regulators.” The argument over what is a fair profit margin for financial transactions goes back centuries, if not millennia.
SOFR[1] itself is an excellent solution to the regulators’ dilemma. It is published for free by the NY Fed based on a robust data set of $800 billion to $1.2 trillion of overnight Treasury lending transactions every day. Through the NY Fed’s consultation process with major market participants (the Alternative Reference Rates Committee or ARRC) SOFR has been the officially designated successor since June 2017. But what’s not to love about SOFR?
- It is an overnight rate. The most accurate version (and the standard version in derivatives fallback rates) is the overnight compounded-in-arrears incarnation of SOFR, which requires lenders to completely retool their operations to calculate interest payments. It’s expensive to remedy this but after some years of trying most vendors claim they can do it. Two alternative ways to reduce the calculation burden are (a) use simple interest SOFR, and (b) refer to a public index of the compounded rate from the NY Fed’s website.
- We do not know the payment amount until two days before the payment date. There is some hope this problem will go away with the appearance of Term SOFR. As banks cease all interbank LIBOR transactions in July 2021, they will start trading in 1-month, 3-month or 6-month estimates of the compounded SOFR rate and start to create liquidity in that incarnation of the index. The Chicago Mercantile Exchange (CME) already uses futures prices to estimate Term SOFR rates. If it exists, Term SOFR is first on the list of fallback rates in many loan contracts. Its presence on Earth among ordinary end-users is overdue.
- SOFR is a true risk-free rate, as safe and more liquid than Treasuries since it represents the borrowing rate for secured transactions. That differs from LIBOR which bakes in both the risk-free level of rates and the credit risk/funding cost of the banking system. In March 2020 SOFR went essentially to zero while LIBOR went above 140 basis points due to the strain banks were having in raising funds. This is more than just “desirable” for money-center banks. It is the natural order of the universe, and from their point of view SOFR is problematic.
To see the bank’s point of view, go back to March 2020. LIBOR is spiking up during the crisis—and therefore so is interest income linked to LIBOR indexed loans, which helps cover the higher cost of defaults among bank borrowers in the hospitality sector. While some borrowers draw down their revolving lines of credit (despite the public signal of desperation this entails), many others are deterred by the higher cost of borrowing. And what happens to existing fixed rate debt? As LIBOR spikes, the fair value of such debt declines, adding some cushion and flexibility in banks’ capital positions.
The bank survives to be what bankers know and love: the conduit from savers to investors, the underwriter/executor of emergency lending facilities, the generous liquidity provider to relationship partners, the builder of our cyber-safe payments infrastructure, the trusty guardian of the store of value and living sign of confidence among all people, which we call “Credit.”
If SOFR were the only game in town, with loans priced to a fixed spread, bank capital would be more volatile. In times like March 2020, interest revenue would fall just when credit spreads—the market indicator of impending defaults—spike up. More clients would draw down their low-cost line of credit during the crisis, straining bank liquidity at the worst time. Without a self-correcting cost of funds index embedded in their loans, much more will be riding on banks getting the spread to that index correct. Increasing uncertainty in the levels of bank equity capital will make lenders less willing to make loans during good times, or step in to help resolve the next crisis by buying troubled lenders. Loan costs and spreads somehow need to go up under SOFR to reconcile lenders’ concerns.
The credit sensitive rates BSBY,[2] IBA Bank Yield Index,[3] Markit CRITR[4] share similar approaches: analyze bank debt transactions in primary markets, incorporate some secondary market or executable quote data, filter and adjust for outliers, and demonstrate a remarkably tight backward-looking fit to LIBOR data. AMERIBOR is derived from deposits and debt traded on a specialized exchange. All of these are based on much less data than SOFR—from $2 billion to $50 billion of transaction data depending on the circumstances. Could the whole world really adopt one or more of these upstart alternatives given their data limitations and the fragility that implies?
To listen to regulators Quarles (Federal Reserve) and Gensler (SEC),[5] “the deniers and laggards are engaged in magical thinking. LIBOR is over.” With reference to a credit sensitive rate they state, “the ARRC does not support more than minimal use of other rates in capital markets or for derivatives, and market participants should not expect such rates to be widely available. Make no mistake: it might look a bit different, but it’s still the same emperor” with no clothes. “It presents similar risks to financial stability and market resiliency” as LIBOR does.
And as any yeoman farmer would appreciate, “If we manage this inflection point well over the next few months, the broad, sunlit uplands of a stronger financial system beckon.”
But … anything is possible. Initially we can expect to see more innovation and experimentation, with some take-up of credit sensitive rates in the bank loan market and real estate lending. The concept of hybrid loans—such as SOFR + a credit spread index—could appeal to all sides but at the expense of more operational complexity. Though a hundred flowers are blooming now, the benefits of standardization will lead to consolidation, in my view. Are the major banks really so central to the universe of credit origination and financial intermediation that they can dictate market practice for the successor rate? Will the credit-sensitive rates morph into a secondary role as an interbank basis swap market, just to accommodate odd lots of loans? Could they turn the tables on SOFR and the ARRC’s rational attempt to control the financial universe, and win this race after all? Stay tuned to find out!
Statements of fact and opinions expressed herein are those of the individual author and are not necessarily those of the Society of Actuaries, the newsletter editors, or the respective author’s employer.
Joe Koltisko, FSA, CFA, is a vice president and chief derivatives officer of New York Life Insurance Company. He can be reached at jkoltisko@nyl.com.
Endnotes
[1]https://www.newyorkfed.org/medialibrary/Microsites/arrc/files/2021/users-guide-to-sofr2021-update.pdf
https://www.newyorkfed.org/markets/reference-rates/sofr
[2]https://www.bloomberg.com/professional/introducing-the-bloomberg-short-term-bank-yield-index-bsby/
[3] https://www.theice.com/iba/bank-yield-index
[4]https://ihsmarkit.com/products/credit-inclusive-benchmarks.html
[5]Fed Vice Chair Quarles https://www.federalreserve.gov/supervisionreg/files/quarles-libor-presentation-20210611.pdf; SEC Chair Gensler https://www.sec.gov/news/public-statement/gensler-fsoc-libor-2021-06-11