This commentary is written by Martin Fridson, a high-yield market veteran who is chief investment officer of Lehmann Livian Fridson Advisors LLC as well as a contributing analyst to S&P Global Market Intelligence.
Lowest-rated HY bonds account for all spread tightening of past 12 months
High-yield investors have truly undergone agony and ecstasy over the past year. It has been a long climb back from the nadir of March 23, 2020, when the ICE BofA US High Yield Index's option-adjusted spread, or OAS, maxed out at +1,087 basis points. At that point, the high-yield market as a whole was priced at a distressed level, based on the definition that I introduced around 1990, i.e., yield spread-versus-Treasurys of +1,000 bps or more.
In January 2021, the market recovery reached a milestone. At month-end, the index's OAS was lower than it was one year earlier. That had not happened since January 2020.
Investors might be wise to temper their celebrating, though. True, the table below shows that the 2021 ICE BofA US High Yield Index's OAS was 17 bps tighter on Feb. 5, 2021, than it was one year earlier. Both the BB and B segments, however, were wider than on the same date in 2020.
In short, all of the year-over-year reduction in the high-yield risk premium was attributable to the CCC & Lower segment. The bottom-rated tier's huge 318 bps OAS reduction just barely offset the other rating categories' much smaller increases because of its low percentage weighting in the overall index. On Feb. 5, 2020, the CCC & Lower group accounted for just 12.19% of the ICE BofA US High Yield Index's market value.
It is not altogether surprising that the CCC & Lower spread contracted dramatically as the Fed effectively backstopped the financial markets, sparking a rally from the March 23, 2020, low point. As the table shows, the CCC & Lower OAS stood at +1,007 bps, while the combined BB+B OAS was +287 bps. According to the methodology introduced in "CCC & Lower relatively cheap in rich HY market," the fair value for the CCC & Lower segment, given the prevailing BB+B spread, was just +745 bps. The CCC & Lower undervaluation of 262 bps exceeded one standard deviation, or 254 bps, our threshold for declaring an extreme misevaluation.
While a large contraction in the CCC & Lower spread made sense in response to the change in conditions, it is striking that BBs and Bs remain wider than they were a year ago. The CCC & Lower spread has now swung to the overvalued side, by our analysis, although the -112-bps gap between the fair value spread of +801 bps and the actual spread of +689 bps does not represent an extreme. A plausible interpretation of these observations is that with Treasury rates currently at depressed levels due to the Fed's quantitative easing, yield-starved investors are willing to accept a narrower-than-appropriate spread on CCC & Lower issues. BBs and Bs, with their lower yields, do not make such tempting targets.
How a pandemic and a recession changed the HY universe's composition
The unforeseen blow that COVID-19 inflicted on the economy dropped a sizable chunk of the investment-grade market to speculative-grade. As detailed in the table below, fallen angels consequently expanded from 9.06% to 15.96% of the high-yield market's face amount during 2020. As a percentage of the ICE BofA US High Yield Index's issues, fallen angels' share ended a steady decline from 14.56% in 2015, to rise to 16.16%, the highest level since 2011's 19.22%.
The large fallen angel influx in 2020 had noteworthy knock-on effects on the high-yield universe's composition. To begin with, the migration of bonds from investment-grade helped boost the BB share of the index's face amount from 47.18% to 54.12%. (Most bonds exiting IG wind up, at least initially, in the BB category). The 54.12% fallen-angel share is the highest proportion since such data became available for the ICE BofA US High Yield Index in 1996. The previous high was 47.34% in 2016.
Note that these numbers all represent net increases in the fallen angel share. In a typical year, some bonds fall from investment-grade to speculative-grade, but others do the opposite. Also affecting the calculations are redemptions, aging out (the ICE BofA US High Yield Index excludes issues with maturities of less than one year), and defaults.
As a result of the shift in ratings mix, portfolio managers who are benchmarked against the ICE BofA US High Yield index have a significantly higher-quality bogey than formerly. As a group, fallen angels also have longer average maturities than original-issue high-yield bonds, since investment-grade companies are able to issue longer-maturity bonds than the noninvestment-grade market will accept. This difference is reflected in 2020's lengthening of the average high-yield maturity from 5.96 years to 6.56 years. That is the longest average maturity since 2013 and the year-over-year change represented the biggest increase since 2002.
Reflecting their origin as IG issues, fallen angels also have lower par-weighted average coupons than original issues. Therefore, the 2020 increase in fallen angel share of HY face amount contributed to a drop in the high-yield index's average par-weighted coupon from 6.32% to 5.99%. The 5.99% figure is the lowest year-end figure in the ICE BofA US High Yield Index's history, noting that the long-run trend has been downward as a function of declining new-issue yields.
The combination of a longer average maturity and lower par-weighted coupon, attributable in substantial measure to the large influx of fallen angels in 2020, contributed to a lengthening of the high-yield index's effective duration. It increased from 3.30 to 3.67. That change partially reversed 2019's drop from 4.19 to 3.30, which was the shortest in the index's available history. Note that the index's effective duration at a point in time is also a function of its prevailing yield, which is independent of the change in the fallen angels' share of the high-yield universe.
Finally, the number of issues in the ICE BofA US High Yield Index grew by 255 in 2020. The increase from 1,775 to 2,030 reversed an uninterrupted downtrend from 2,314 in 2014 but left the count 12.27% below its peak in that year. Last year's net addition of 255 issues (after accounting for defaults, redemptions, upgrades to IG, and aging out) was the largest since the index grew by 440 bonds in 2002. Note, though, that 2002's figure represented a 32.25% year-over-year increase versus just 14.37% in 2020. Original issues accounted for almost as much of the year's net increase (120 bonds) as fallen angels (135 bonds). Fallen angels, however, represent a far smaller component of the high-yield index than original issues. Due to their much smaller base, fallen angels' 2020 net addition of issues represented a 69.95% increase versus just 7.59% for original issues.
For the benefit of investors who are benchmarked against the ICE BofA US High Yield Constrained Index, the table below details the 2020-versus-2019 change in ratings mix. As with the ICE BofA US High Yield Index, the BB share of face amount grew at the expense of the other rating categories. At 53.50% as of Dec. 31, 2020, the top rating category had a slightly smaller share than was the case for the unconstrained index, at 54.12%.
Neutral weighting recommended for leveraged loans
High-yield bonds and leveraged loans are currently very closely aligned in relative valuation terms. As illustrated in the chart below, the latest update of our monthly loans-versus-bonds relative value analysis shows loans to be just a touch expensive vis-à-vis bonds. At -0.38 on the vertical scale, loans on Jan. 31, 2021, were far from the extreme valuation (minus one standard deviation) that would trigger a recommendation to underweight them within a combined bond-loan portfolio. Accordingly, a neutral weighting on loans is currently recommended. On Dec. 31, 2020, the reading stood at 0.53.
Details of our relative valuation methodology for the two categories of debt appeared in "Loans vs. bonds – Determining relative value." In brief, we compare the three-year discounted spread on the S&P/LSTA Leveraged Loan Index and the option-adjusted spread on the ICE BofA U.S. High Yield Index, after first adjusting for differences in ratings mix between the two asset classes. We convert the difference in these spreads into an index geared to one standard deviation from the mean in either direction. A reading of plus 1.0 indicates that bonds are extremely rich versus loans, and a reading of negative 1.0 indicates that loans are extremely rich versus bonds. Inside those bands, we recommend neutral weightings of loans and bonds.
Research assistance by Bach Ho and Ducheng Peng.
ICE BofA Index System data is used by permission. Copyright © 2020 ICE Data Services. The use of the above in no way implies that ICE Data Services or any of its affiliates endorses the views or interpretation or the use of such information or acts as any endorsement of Lehmann Livian Fridson Advisors LLC's use of such information. The information is provided "as is" and none of ICE Data Services or any of its affiliates warrants the accuracy or completeness of the information.