By the Numbers
Finding fair value in the LIBOR-to-SOFR transition
Steven Abrahams and Caroline Chen | January 28, 2022
This document is intended for institutional investors and is not subject to all of the independence and disclosure standards applicable to debt research reports prepared for retail investors.
The CLO market has started to transition away from LIBOR with 12 deals backed by broadly syndicated loans pricing to SOFR so far this year, but the debate continues about the fair margin over the new index. New ‘AAA’ classes from regular issuers came to market in December at spreads to 3-month LIBOR between 108 bp and 121 bp, and in January at spreads to 3-month SOFR between 130 bp and 152 bp. Based on projected CLO equity yield for at least one benchmark deal, current SOFR spreads look equity friendly compared to original LIBOR levels and compared to hypothetical current LIBOR levels at the wide end of market pricing. Compared to estimated mid-market or tight levels on a hypothetical LIBOR deal issued today, current SOFR spreads look more debt friendly.
The approach
The market ultimately will decide where to price SOFR-indexed CLO debt, but one angle on fair value looks at equity internal rate of return. Assuming the market sees both LIBOR and SOFR as practically the same—more on that later—equity IRR signals whether deal execution tilts equity or debt friendly. Since LIBOR is no longer an option for indexing CLO debt, it is a hypothetical exercise but still useful to see if SOFR at this point has favored one form of liability over the other.
Neuberger Berman has recently issued similar deals pricing to both LIBOR and SOFR. Neuberger Berman Loan Advisers CLO 45 priced on October 7, 2021, indexed to LIBOR, and Neuberger Berman Loan Advisers CLO 47 priced on January 21, 2022, indexed to SOFR. The exercise takes NEUB 45 and estimates where it would price at new issue today using LIBOR. There is some uncertainty about exact levels, so the analysis assumes levels for a wide, mid-market and tight execution. The exercise also estimates where NEUB 45 would price at new issue today using SOFR, with SOFR levels taken directly from NEUB 47. Each hypothetical execution leads to an equity IRR. And comparing equity IRR to one another shows whether the deals are relatively more equity or debt friendly.
It is important to note that even a LIBOR deal hypothetically issued today will transition to SOFR after June 30, 2023. The Alternative Reference Rate Committee under the Fed has recommended that LIBOR-indexed debt transition to SOFR after that date by adding a spread adjustment (Exhibit 1). The analysis consequently assumes that the debt on each hypothetical LIBOR deal priced today will transition to SOFR as recommended when its coupon resets in August 2023.
Exhibit 1: Transition to SOFR often involves an ARRC spread adjustment
Source: ARRC, Amherst Pierpont Securities
Finally, the analysis also makes assumptions about transition to SOFR in the loans backing the hypothetically repriced NEUB 45. The loans will likely transition continually as the portfolio reinvests in new loans and as borrowers amend loans to reference SOFR. This transition could follow an unlimited number of paths, but the assumption here is that a steady share of loans transitions each month until every loan has transitioned by July 2023. The current spread between 3-month LIBOR and SOFR is 10 bp, for example, but forward rates provided by Intex suggest it eventually increases to 22 bp. It is reasonable to assume that the spread adjustment to SOFR increases a little each month. The same assumption holds for other LIBOR-to-SOFR indices.
The analysis also assumes prepayments of 20 CPR, defaults of 2 CDR, recoveries on default of 70% and delay between default and recover of six months.
Results
The current relative value of SOFR execution for CLO equity or debt obviously depends on where a hypothetical current LIBOR deal would price. The results for projected equity IRR under assumptions made here if NEUB 45 repriced today (Exhibit 2):
- At original LIBOR spreads: 15.29%
- At wide LIBOR spreads: 15.12%
- At mid-market LIBOR spreads: 18.57%
- At tight LIBOR spreads: 19.34%
- At market SOFR spreads: 16.42%
Exhibit 2: Hypothetical NEUB 45 executions and projected equity IRR
Source: LCD, Amherst Pierpont Securities
For investors that assume hypothetical new LIBOR deals would price either at levels from late 2021 or at the wide end of the current estimated LIBOR range, current SOFR levels deliver a higher IRR to equity and look equity friendly. For investors that assume hypothetical new LIBOR deals would come at tighter levels than late 2021, then current SOFR deals deliver a lower equity IRR and look debt friendly.
Debt spreads should be even wider
One assumption made in this analysis is that LIBOR and SOFR indices are practically interchangeable, but that is almost certainly not the case. In a crisis, such as March 2020, LIBOR or other credit-sensitive rates will almost certainly widen to riskless rates and possibly go up. SOFR, on the other hand, as a riskless rate will likely go down in a flight-to-quality. A LIBOR-indexed CLO coupon would consequently go up while a SOFR-indexed coupon would go down. SOFR-indexed debt where the coupon could go down in a crisis should command some current spread compensation. The market will likely price for that only after the next crisis shows the impact of indexing off a riskless rate.